Industry leading face manipulation platform
3.7.0 (#1175)
Henry Ruhsa2cbfd73b101
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.ymlindex 829cc9be..59b7e188 100644--- a/.github/workflows/ci.yml+++ b/.github/workflows/ci.yml@@ -33,7 +33,7 @@ jobs:uses: actions/setup-python@v5with:python-version: '3.12'- - run: python install.py --onnxruntime default --skip-conda+ - run: python install.py default --skip-conda- run: pip install pytest- run: pytestreport:@@ -48,7 +48,7 @@ jobs:uses: actions/setup-python@v5with:python-version: '3.12'- - run: python install.py --onnxruntime default --skip-conda+ - run: python install.py default --skip-conda- run: pip install coveralls- run: pip install pytest- run: pip install pytest-covdiff --git a/facefusion.ini b/facefusion.iniindex f1518a3b..171b86b4 100644--- a/facefusion.ini+++ b/facefusion.ini@@ -32,6 +32,9 @@ reference_face_position =reference_face_distance =reference_frame_number =+[face_tracker]+face_tracker_score =+[face_masker]face_occluder_model =face_parser_model =@@ -50,6 +53,9 @@ trim_frame_end =temp_frame_format =keep_temp =+[frame_process]+target_frame_amount =+[output_creation]output_image_quality =output_image_scale =@@ -125,7 +131,6 @@ execution_thread_count =[memory]video_memory_strategy =-system_memory_limit =[misc]log_level =diff --git a/facefusion/app_context.py b/facefusion/app_context.pyindex d54f961e..8f7d404c 100644--- a/facefusion/app_context.py+++ b/facefusion/app_context.py@@ -5,12 +5,14 @@ from facefusion.types import AppContextdef detect_app_context() -> AppContext:+ jobs_path = os.path.join('facefusion', 'jobs')+ uis_path = os.path.join('facefusion', 'uis')frame = sys._getframe(1)while frame:- if os.path.join('facefusion', 'jobs') in frame.f_code.co_filename:+ if jobs_path in frame.f_code.co_filename:return 'cli'- if os.path.join('facefusion', 'uis') in frame.f_code.co_filename:+ if uis_path in frame.f_code.co_filename:return 'ui'frame = frame.f_backreturn 'cli'diff --git a/facefusion/args.py b/facefusion/args.pyindex c5d1ea24..06adfc57 100644--- a/facefusion/args.py+++ b/facefusion/args.py@@ -33,6 +33,7 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('reference_face_position', args.get('reference_face_position'))apply_state_item('reference_face_distance', args.get('reference_face_distance'))apply_state_item('reference_frame_number', args.get('reference_frame_number'))+ apply_state_item('face_tracker_score', args.get('face_tracker_score'))apply_state_item('face_occluder_model', args.get('face_occluder_model'))apply_state_item('face_parser_model', args.get('face_parser_model'))apply_state_item('face_mask_types', args.get('face_mask_types'))@@ -45,6 +46,7 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('trim_frame_end', args.get('trim_frame_end'))apply_state_item('temp_frame_format', args.get('temp_frame_format'))apply_state_item('keep_temp', args.get('keep_temp'))+ apply_state_item('target_frame_amount', args.get('target_frame_amount'))apply_state_item('output_image_quality', args.get('output_image_quality'))apply_state_item('output_image_scale', args.get('output_image_scale'))apply_state_item('output_audio_encoder', args.get('output_audio_encoder'))@@ -77,7 +79,6 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('benchmark_resolutions', args.get('benchmark_resolutions'))apply_state_item('benchmark_cycle_count', args.get('benchmark_cycle_count'))apply_state_item('video_memory_strategy', args.get('video_memory_strategy'))- apply_state_item('system_memory_limit', args.get('system_memory_limit'))apply_state_item('log_level', args.get('log_level'))apply_state_item('halt_on_error', args.get('halt_on_error'))apply_state_item('job_id', args.get('job_id'))diff --git a/facefusion/benchmarker.py b/facefusion/benchmarker.pyindex 2f404886..4ba16bd4 100644--- a/facefusion/benchmarker.py+++ b/facefusion/benchmarker.py@@ -9,7 +9,7 @@ import facefusion.choicesfrom facefusion import content_analyser, core, state_managerfrom facefusion.cli_helper import render_tablefrom facefusion.download import conditional_download, resolve_download_url-from facefusion.face_store import clear_static_faces+from facefusion.face_store import clear_facesfrom facefusion.filesystem import get_file_extensionfrom facefusion.types import BenchmarkCycleSetfrom facefusion.vision import count_video_frame_total, detect_video_fps@@ -64,7 +64,7 @@ def cycle(cycle_count : int) -> BenchmarkCycleSet:if state_manager.get_item('benchmark_mode') == 'cold':content_analyser.analyse_image.cache_clear()content_analyser.analyse_video.cache_clear()- clear_static_faces()+ clear_faces()start_time = perf_counter()core.conditional_process()diff --git a/facefusion/choices.py b/facefusion/choices.pyindex 374ce639..a747a574 100755--- a/facefusion/choices.py+++ b/facefusion/choices.py@@ -2,7 +2,7 @@ import loggingfrom typing import List, Sequence, get_argsfrom facefusion.common_helper import create_float_range, create_int_range-from facefusion.types import Angle, AudioEncoder, AudioFormat, AudioTypeSet, BenchmarkMode, BenchmarkResolution, BenchmarkSet, DownloadProvider, DownloadProviderSet, DownloadScope, EncoderSet, ExecutionProvider, ExecutionProviderSet, FaceDetectorModel, FaceDetectorSet, FaceLandmarkerModel, FaceMaskArea, FaceMaskAreaSet, FaceMaskRegion, FaceMaskRegionSet, FaceMaskType, FaceOccluderModel, FaceParserModel, FaceSelectorMode, FaceSelectorOrder, Gender, ImageFormat, ImageTypeSet, JobStatus, LogLevel, LogLevelSet, Race, Score, TempFrameFormat, UiWorkflow, VideoEncoder, VideoFormat, VideoMemoryStrategy, VideoPreset, VideoTypeSet, VoiceExtractorModel+from facefusion.types import Angle, AudioEncoder, AudioFormat, AudioTypeSet, BenchmarkMode, BenchmarkResolution, BenchmarkSet, DownloadProvider, DownloadProviderSet, DownloadScope, EncoderSet, ExecutionProvider, ExecutionProviderSet, FaceDetectorModel, FaceDetectorSet, FaceLandmarkerModel, FaceMaskArea, FaceMaskAreaSet, FaceMaskRegion, FaceMaskRegionSet, FaceMaskType, FaceOccluderModel, FaceParserModel, FaceSelectorGender, FaceSelectorMode, FaceSelectorOrder, FaceSelectorRace, Gender, ImageFormat, ImageTypeSet, JobStatus, LogLevel, LogLevelSet, Race, Score, TempFrameFormat, UiWorkflow, VideoEncoder, VideoFormat, VideoMemoryStrategy, VideoPreset, VideoTypeSet, VoiceExtractorModelface_detector_set : FaceDetectorSet =\{@@ -16,8 +16,10 @@ face_detector_models : List[FaceDetectorModel] = list(get_args(FaceDetectorModelface_landmarker_models : List[FaceLandmarkerModel] = list(get_args(FaceLandmarkerModel))face_selector_modes : List[FaceSelectorMode] = list(get_args(FaceSelectorMode))face_selector_orders : List[FaceSelectorOrder] = list(get_args(FaceSelectorOrder))-face_selector_genders : List[Gender] = list(get_args(Gender))-face_selector_races : List[Race] = list(get_args(Race))+genders : List[Gender] = list(get_args(Gender))+races : List[Race] = list(get_args(Race))+face_selector_genders : List[FaceSelectorGender] = list(get_args(FaceSelectorGender))+face_selector_races : List[FaceSelectorRace] = list(get_args(FaceSelectorRace))face_occluder_models : List[FaceOccluderModel] = list(get_args(FaceOccluderModel))face_parser_models : List[FaceParserModel] = list(get_args(FaceParserModel))face_mask_types : List[FaceMaskType] = list(get_args(FaceMaskType))@@ -153,7 +155,6 @@ job_statuses : List[JobStatus] = list(get_args(JobStatus))benchmark_cycle_count_range : Sequence[int] = create_int_range(1, 10, 1)execution_thread_count_range : Sequence[int] = create_int_range(1, 32, 1)-system_memory_limit_range : Sequence[int] = create_int_range(0, 128, 4)face_detector_margin_range : Sequence[int] = create_int_range(0, 100, 1)face_detector_angles : Sequence[Angle] = create_int_range(0, 270, 90)face_detector_score_range : Sequence[Score] = create_float_range(0.0, 1.0, 0.05)@@ -162,6 +163,8 @@ face_mask_blur_range : Sequence[float] = create_float_range(0.0, 1.0, 0.05)face_mask_padding_range : Sequence[int] = create_int_range(0, 100, 1)face_selector_age_range : Sequence[int] = create_int_range(0, 100, 1)reference_face_distance_range : Sequence[float] = create_float_range(0.0, 1.0, 0.05)+face_tracker_score_range : Sequence[Score] = create_float_range(0.0, 0.5, 0.05)+target_frame_amount_range : Sequence[int] = create_int_range(0, 10, 1)output_image_quality_range : Sequence[int] = create_int_range(0, 100, 1)output_image_scale_range : Sequence[float] = create_float_range(0.25, 8.0, 0.25)output_audio_quality_range : Sequence[int] = create_int_range(0, 100, 1)diff --git a/facefusion/common_helper.py b/facefusion/common_helper.pyindex 3587784d..05e4c4e9 100644--- a/facefusion/common_helper.py+++ b/facefusion/common_helper.py@@ -78,6 +78,12 @@ def get_first(__list__ : Any) -> Any:return None+def get_middle(__list__ : Any) -> Any:+ if isinstance(__list__, Sequence) and __list__:+ return __list__[len(__list__) // 2]+ return None++def get_last(__list__ : Any) -> Any:if isinstance(__list__, Reversible):return next(reversed(__list__), None)diff --git a/facefusion/config.py b/facefusion/config.pyindex e8e307ce..63aab509 100644--- a/facefusion/config.py+++ b/facefusion/config.py@@ -1,29 +1,20 @@from configparser import ConfigParser+from functools import lru_cachefrom typing import List, Optionalfrom facefusion import state_managerfrom facefusion.common_helper import cast_bool, cast_float, cast_int-CONFIG_PARSER = None--def get_config_parser() -> ConfigParser:- global CONFIG_PARSER-- if CONFIG_PARSER is None:- CONFIG_PARSER = ConfigParser()- CONFIG_PARSER.read(state_manager.get_item('config_path'), encoding = 'utf-8')- return CONFIG_PARSER---def clear_config_parser() -> None:- global CONFIG_PARSER-- CONFIG_PARSER = None+@lru_cache+def get_static_config_parser() -> ConfigParser:+ config_parser = ConfigParser()+ config_parser.read(state_manager.get_item('config_path'), encoding = 'utf-8')+ return config_parserdef get_str_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[str]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return config_parser.get(section, option)@@ -31,7 +22,7 @@ def get_str_value(section : str, option : str, fallback : Optional[str] = None)def get_int_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[int]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return config_parser.getint(section, option)@@ -39,7 +30,7 @@ def get_int_value(section : str, option : str, fallback : Optional[str] = None)def get_float_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[float]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return config_parser.getfloat(section, option)@@ -47,7 +38,7 @@ def get_float_value(section : str, option : str, fallback : Optional[str] = Nonedef get_bool_value(section : str, option : str, fallback : Optional[str] = None) -> Optional[bool]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return config_parser.getboolean(section, option)@@ -55,7 +46,7 @@ def get_bool_value(section : str, option : str, fallback : Optional[str] = None)def get_str_list(section : str, option : str, fallback : Optional[str] = None) -> Optional[List[str]]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return config_parser.get(section, option).split()@@ -65,7 +56,7 @@ def get_str_list(section : str, option : str, fallback : Optional[str] = None) -def get_int_list(section : str, option : str, fallback : Optional[str] = None) -> Optional[List[int]]:- config_parser = get_config_parser()+ config_parser = get_static_config_parser()if config_parser.has_option(section, option) and config_parser.get(section, option).strip():return list(map(int, config_parser.get(section, option).split()))diff --git a/facefusion/content_analyser.py b/facefusion/content_analyser.pyindex 655ed5d8..cf21f06f 100644--- a/facefusion/content_analyser.py+++ b/facefusion/content_analyser.py@@ -1,16 +1,14 @@from functools import lru_cache-from typing import List, Tuple+from typing import Tupleimport numpyfrom tqdm import tqdmfrom facefusion import inference_manager, state_manager, translator-from facefusion.common_helper import is_macosfrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.execution import has_execution_providerfrom facefusion.filesystem import resolve_relative_pathfrom facefusion.thread_helper import conditional_thread_semaphore-from facefusion.types import Detection, DownloadScope, DownloadSet, ExecutionProvider, Fps, InferencePool, ModelSet, VisionFrame+from facefusion.types import Detection, DownloadScope, DownloadSet, Fps, InferencePool, ModelSet, VisionFramefrom facefusion.vision import detect_video_fps, fit_contain_frame, read_image, read_video_frameSTREAM_COUNTER = 0@@ -119,12 +117,6 @@ def clear_inference_pool() -> None:inference_manager.clear_inference_pool(__name__, model_names)-def resolve_execution_providers() -> List[ExecutionProvider]:- if is_macos() and has_execution_provider('coreml'):- return [ 'cpu' ]- return state_manager.get_item('execution_providers')--def collect_model_downloads() -> Tuple[DownloadSet, DownloadSet]:model_set = create_static_model_set('full')model_hash_set = {}@@ -175,10 +167,12 @@ def analyse_video(video_path : str, trim_frame_start : int, trim_frame_end : intfor frame_number in frame_range:if frame_number % int(video_fps) == 0:vision_frame = read_video_frame(video_path, frame_number)- total += 1- if analyse_frame(vision_frame):- counter += 1+ if numpy.any(vision_frame):+ total += 1++ if analyse_frame(vision_frame):+ counter += 1if counter > 0 and total > 0:rate = counter / total * 100diff --git a/facefusion/core.py b/facefusion/core.pyindex 1feb1d21..c8168c39 100755--- a/facefusion/core.py+++ b/facefusion/core.py@@ -5,14 +5,13 @@ import signalimport sysfrom time import time-from facefusion import benchmarker, cli_helper, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, hash_helper, logger, state_manager, translator, voice_extractor+from facefusion import benchmarker, cli_helper, content_analyser, hash_helper, logger, state_manager, translatorfrom facefusion.args import apply_args, collect_job_args, reduce_job_args, reduce_step_argsfrom facefusion.download import conditional_download_hashes, conditional_download_sourcesfrom facefusion.exit_helper import hard_exit, signal_exitfrom facefusion.filesystem import get_file_extension, get_file_name, is_image, is_video, resolve_file_paths, resolve_file_patternfrom facefusion.jobs import job_helper, job_manager, job_runnerfrom facefusion.jobs.job_list import compose_job_list-from facefusion.memory import limit_system_memoryfrom facefusion.processors.core import get_processors_modulesfrom facefusion.program import create_programfrom facefusion.program_helper import validate_args@@ -41,11 +40,6 @@ def cli() -> None:def route(args : Args) -> None:- system_memory_limit = state_manager.get_item('system_memory_limit')-- if system_memory_limit and system_memory_limit > 0:- limit_system_memory(system_memory_limit)-if state_manager.get_item('command') == 'force-download':error_code = force_download()hard_exit(error_code)@@ -107,21 +101,9 @@ def pre_check() -> bool:def common_pre_check() -> bool:- common_modules =\- [- content_analyser,- face_classifier,- face_detector,- face_landmarker,- face_masker,- face_recognizer,- voice_extractor- ]-content_analyser_content = inspect.getsource(content_analyser).encode()- content_analyser_hash = hash_helper.create_hash(content_analyser_content)- return all(module.pre_check() for module in common_modules) and content_analyser_hash == 'b14e7b92'+ return hash_helper.create_hash(content_analyser_content) == '975d67d6'def processors_pre_check() -> bool:@@ -132,22 +114,19 @@ def processors_pre_check() -> bool:def force_download() -> ErrorCode:- common_modules =\- [- content_analyser,- face_classifier,- face_detector,- face_landmarker,- face_masker,- face_recognizer,- voice_extractor- ]+ download_scope = state_manager.get_item('download_scope')available_processors = [ get_file_name(file_path) for file_path in resolve_file_paths('facefusion/processors/modules') ]processor_modules = get_processors_modules(available_processors)+ common_modules = []++ for processor_module in processor_modules:+ for common_module in processor_module.get_common_modules():+ if common_module not in common_modules:+ common_modules.append(common_module)for module in common_modules + processor_modules:if hasattr(module, 'create_static_model_set'):- for model in module.create_static_model_set(state_manager.get_item('download_scope')).values():+ for model in module.create_static_model_set(download_scope).values():model_hash_set = model.get('hashes')model_source_set = model.get('sources')diff --git a/facefusion/face_analyser.py b/facefusion/face_creator.pysimilarity index 55%rename from facefusion/face_analyser.pyrename to facefusion/face_creator.pyindex 76b9b621..fb4a8a81 100644--- a/facefusion/face_analyser.py+++ b/facefusion/face_creator.py@@ -2,14 +2,13 @@ from typing import List, Optionalimport numpy-from facefusion import state_manager-from facefusion.common_helper import get_first+from facefusion import face_store, state_manager+from facefusion.common_helper import get_first, get_middlefrom facefusion.face_classifier import classify_facefrom facefusion.face_detector import detect_faces, detect_faces_by_angle-from facefusion.face_helper import apply_nms, convert_to_face_landmark_5, estimate_face_angle, get_nms_threshold+from facefusion.face_helper import apply_nms, average_points, convert_to_face_landmark_5, estimate_face_angle, get_nms_thresholdfrom facefusion.face_landmarker import detect_face_landmark, estimate_face_landmark_68_5from facefusion.face_recognizer import calculate_face_embedding-from facefusion.face_store import get_static_faces, set_static_facesfrom facefusion.types import BoundingBox, Face, FaceLandmark5, FaceLandmarkSet, FaceScoreSet, Score, VisionFrame@@ -47,7 +46,9 @@ def create_faces(vision_frame : VisionFrame, bounding_boxes : List[BoundingBox],}face_embedding, face_embedding_norm = calculate_face_embedding(vision_frame, face_landmark_set.get('5/68'))gender, age, race = classify_face(vision_frame, face_landmark_set.get('5/68'))+faces.append(Face(+ origin = 'detect',bounding_box = bounding_box,score_set = face_score_set,landmark_set = face_landmark_set,@@ -68,7 +69,102 @@ def get_one_face(faces : List[Face], position : int = 0) -> Optional[Face]:return None-def get_average_face(faces : List[Face]) -> Optional[Face]:+def get_many_faces(vision_frames : List[VisionFrame]) -> List[Face]:+ many_faces : List[Face] = []++ for vision_frame in vision_frames:+ if numpy.any(vision_frame):+ all_bounding_boxes = []+ all_face_scores = []+ all_face_landmarks_5 = []++ for face_detector_angle in state_manager.get_item('face_detector_angles'):+ if face_detector_angle == 0:+ bounding_boxes, face_scores, face_landmarks_5 = detect_faces(vision_frame)+ else:+ bounding_boxes, face_scores, face_landmarks_5 = detect_faces_by_angle(vision_frame, face_detector_angle)+ all_bounding_boxes.extend(bounding_boxes)+ all_face_scores.extend(face_scores)+ all_face_landmarks_5.extend(face_landmarks_5)++ if all_bounding_boxes and all_face_scores and all_face_landmarks_5 and state_manager.get_item('face_detector_score') > 0:+ faces = create_faces(vision_frame, all_bounding_boxes, all_face_scores, all_face_landmarks_5)++ if faces:+ many_faces.extend(faces)++ return many_faces+++def get_static_faces(vision_frames : List[VisionFrame]) -> List[Face]:+ many_faces : List[Face] = []++ for vision_frame in vision_frames:+ faces = face_store.get_faces(vision_frame)++ if not faces:+ with face_store.resolve_lock(vision_frame):+ faces = face_store.get_faces(vision_frame)++ if not faces:+ faces = get_many_faces([ vision_frame ])++ if faces:+ face_store.set_faces(vision_frame, faces)++ many_faces.extend(faces)++ return many_faces+++def refill_faces(faces : List[Optional[Face]]) -> List[Face]:+ fill_faces = []+ anchor_index_previous = -1++ for index, face in enumerate(faces):+ if face:+ for gap_index in range(anchor_index_previous + 1, index):+ average_factor = (gap_index - anchor_index_previous) / (index - anchor_index_previous)+ average_face = average_face_geometry([faces[anchor_index_previous], face], average_factor)+ fill_faces.append(average_face)++ fill_faces.append(face)+ anchor_index_previous = index++ return fill_faces+++def average_face_geometry(faces : List[Face], average_factor : float) -> Face:+ face_first = get_first(faces)+ face_middle = get_middle(faces)+ face_anchor = face_middle++ if average_factor < 0.5:+ face_anchor = face_first++ landmark_set : FaceLandmarkSet =\+ {+ '5': average_points(face_first.landmark_set.get('5'), face_middle.landmark_set.get('5'), average_factor),+ '5/68': average_points(face_first.landmark_set.get('5/68'), face_middle.landmark_set.get('5/68'), average_factor),+ '68': average_points(face_first.landmark_set.get('68'), face_middle.landmark_set.get('68'), average_factor),+ '68/5': average_points(face_first.landmark_set.get('68/5'), face_middle.landmark_set.get('68/5'), average_factor)+ }++ return Face(+ origin = 'refill',+ bounding_box = average_points(face_first.bounding_box, face_middle.bounding_box, average_factor),+ score_set = face_anchor.score_set,+ landmark_set = landmark_set,+ angle = estimate_face_angle(landmark_set.get('68/5')),+ embedding = face_anchor.embedding,+ embedding_norm = face_anchor.embedding_norm,+ gender = face_anchor.gender,+ age = face_anchor.age,+ race = face_anchor.race+ )+++def average_face_identity(faces : List[Face]) -> Optional[Face]:face_embeddings = []face_embeddings_norm = []@@ -80,6 +176,7 @@ def get_average_face(faces : List[Face]) -> Optional[Face]:face_embeddings_norm.append(face.embedding_norm)return Face(+ origin = first_face.origin,bounding_box = first_face.bounding_box,score_set = first_face.score_set,landmark_set = first_face.landmark_set,@@ -93,37 +190,6 @@ def get_average_face(faces : List[Face]) -> Optional[Face]:return None-def get_many_faces(vision_frames : List[VisionFrame]) -> List[Face]:- many_faces : List[Face] = []-- for vision_frame in vision_frames:- if numpy.any(vision_frame):- static_faces = get_static_faces(vision_frame)- if static_faces:- many_faces.extend(static_faces)- else:- all_bounding_boxes = []- all_face_scores = []- all_face_landmarks_5 = []-- for face_detector_angle in state_manager.get_item('face_detector_angles'):- if face_detector_angle == 0:- bounding_boxes, face_scores, face_landmarks_5 = detect_faces(vision_frame)- else:- bounding_boxes, face_scores, face_landmarks_5 = detect_faces_by_angle(vision_frame, face_detector_angle)- all_bounding_boxes.extend(bounding_boxes)- all_face_scores.extend(face_scores)- all_face_landmarks_5.extend(face_landmarks_5)-- if all_bounding_boxes and all_face_scores and all_face_landmarks_5 and state_manager.get_item('face_detector_score') > 0:- faces = create_faces(vision_frame, all_bounding_boxes, all_face_scores, all_face_landmarks_5)-- if faces:- many_faces.extend(faces)- set_static_faces(vision_frame, faces)- return many_faces--def scale_face(target_face : Face, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Face:scale_x = temp_vision_frame.shape[1] / target_vision_frame.shape[1]scale_y = temp_vision_frame.shape[0] / target_vision_frame.shape[0]diff --git a/facefusion/face_helper.py b/facefusion/face_helper.pyindex 553b3e65..aaaca95d 100644--- a/facefusion/face_helper.py+++ b/facefusion/face_helper.py@@ -254,3 +254,23 @@ def merge_matrix(temp_matrices : List[Matrix]) -> Matrix:matrix = numpy.dot(temp_matrix, matrix)return matrix[:2, :]+++def calculate_bounding_box_overlap(bounding_box_a : BoundingBox, bounding_box_b : BoundingBox) -> float:+ intersection_x1 = max(bounding_box_a[0], bounding_box_b[0])+ intersection_y1 = max(bounding_box_a[1], bounding_box_b[1])+ intersection_x2 = min(bounding_box_a[2], bounding_box_b[2])+ intersection_y2 = min(bounding_box_a[3], bounding_box_b[3])+ intersection = max(0, intersection_x2 - intersection_x1) * max(0, intersection_y2 - intersection_y1)+ bounding_box_area = (bounding_box_a[2] - bounding_box_a[0]) * (bounding_box_a[3] - bounding_box_a[1])+ reference_bounding_box_area = (bounding_box_b[2] - bounding_box_b[0]) * (bounding_box_b[3] - bounding_box_b[1])+ union = bounding_box_area + reference_bounding_box_area - intersection++ if union > 0:+ return intersection / union++ return 0.0+++def average_points(points_previous : Points, points_next : Points, average_factor : float) -> Points:+ return points_previous * (1 - average_factor) + points_next * average_factordiff --git a/facefusion/face_selector.py b/facefusion/face_selector.pyindex 00459e9a..a20761cc 100644--- a/facefusion/face_selector.py+++ b/facefusion/face_selector.py@@ -2,26 +2,35 @@ from typing import Listimport numpy+import facefusion.choicesfrom facefusion import state_manager-from facefusion.face_analyser import get_many_faces, get_one_face+from facefusion.common_helper import get_first, get_middle+from facefusion.face_creator import get_one_face, get_static_faces+from facefusion.face_tracker import track_facesfrom facefusion.types import Face, FaceSelectorOrder, Gender, Race, Score, VisionFrame-def select_faces(reference_vision_frame : VisionFrame, target_vision_frame : VisionFrame) -> List[Face]:- target_faces = get_many_faces([ target_vision_frame ])+def select_faces(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], target_vision_frames : List[VisionFrame]) -> List[Face]:+ source_faces = get_static_faces(source_vision_frames)++ if state_manager.get_item('face_tracker_score') > 0:+ target_faces = track_faces(target_vision_frames, state_manager.get_item('face_tracker_score'))+ else:+ target_faces = get_static_faces([ get_middle(target_vision_frames) ])if state_manager.get_item('face_selector_mode') == 'many':- return sort_and_filter_faces(target_faces)+ return sort_and_filter_faces(source_faces, target_faces)if state_manager.get_item('face_selector_mode') == 'one':- target_face = get_one_face(sort_and_filter_faces(target_faces))+ target_face = get_one_face(sort_and_filter_faces(source_faces, target_faces))if target_face:return [ target_face ]if state_manager.get_item('face_selector_mode') == 'reference':- reference_faces = get_many_faces([ reference_vision_frame ])- reference_faces = sort_and_filter_faces(reference_faces)+ reference_faces = get_static_faces([ reference_vision_frame ])+ reference_faces = sort_and_filter_faces(source_faces, reference_faces)reference_face = get_one_face(reference_faces, state_manager.get_item('reference_face_position'))+if reference_face:match_faces = find_match_faces([ reference_face ], target_faces, state_manager.get_item('reference_face_distance'))return match_faces@@ -53,17 +62,33 @@ def calculate_face_distance(face : Face, reference_face : Face) -> float:return 0-def sort_and_filter_faces(faces : List[Face]) -> List[Face]:- if faces:+def sort_and_filter_faces(source_faces : List[Face], target_faces : List[Face]) -> List[Face]:+ if target_faces:if state_manager.get_item('face_selector_order'):- faces = sort_faces_by_order(faces, state_manager.get_item('face_selector_order'))- if state_manager.get_item('face_selector_gender'):- faces = filter_faces_by_gender(faces, state_manager.get_item('face_selector_gender'))- if state_manager.get_item('face_selector_race'):- faces = filter_faces_by_race(faces, state_manager.get_item('face_selector_race'))+ target_faces = sort_faces_by_order(target_faces, state_manager.get_item('face_selector_order'))++ face_selector_gender = state_manager.get_item('face_selector_gender')+ face_selector_race = state_manager.get_item('face_selector_race')++ if source_faces and face_selector_gender == 'auto' or face_selector_race == 'auto':+ source_face = get_first(sort_faces_by_order(source_faces, 'large-small'))++ if source_face:+ if face_selector_gender == 'auto':+ face_selector_gender = source_face.gender+ if face_selector_race == 'auto':+ face_selector_race = source_face.race++ if face_selector_gender in facefusion.choices.genders:+ target_faces = filter_faces_by_gender(target_faces, face_selector_gender)++ if face_selector_race in facefusion.choices.races:+ target_faces = filter_faces_by_race(target_faces, face_selector_race)+if state_manager.get_item('face_selector_age_start') or state_manager.get_item('face_selector_age_end'):- faces = filter_faces_by_age(faces, state_manager.get_item('face_selector_age_start'), state_manager.get_item('face_selector_age_end'))- return faces+ target_faces = filter_faces_by_age(target_faces, state_manager.get_item('face_selector_age_start'), state_manager.get_item('face_selector_age_end'))++ return target_facesdef sort_faces_by_order(faces : List[Face], order : FaceSelectorOrder) -> List[Face]:diff --git a/facefusion/face_store.py b/facefusion/face_store.pyindex bd14caa4..022210a6 100644--- a/facefusion/face_store.py+++ b/facefusion/face_store.py@@ -1,28 +1,42 @@+import threadingfrom typing import List, Optional+import numpy+from facefusion.hash_helper import create_hashfrom facefusion.types import Face, FaceStore, VisionFrame-FACE_STORE : FaceStore =\-{- 'static_faces': {}-}+FACE_STORE : FaceStore = {}+++def get_faces(vision_frame : VisionFrame) -> Optional[List[Face]]:+ if numpy.any(vision_frame):+ vision_hash = create_hash(vision_frame.tobytes())+ if FACE_STORE.get(vision_hash):+ return FACE_STORE.get(vision_hash).get('faces')-def get_face_store() -> FaceStore:- return FACE_STORE+ return None-def get_static_faces(vision_frame : VisionFrame) -> Optional[List[Face]]:- vision_hash = create_hash(vision_frame.tobytes())- return FACE_STORE.get('static_faces').get(vision_hash)+def set_faces(vision_frame : VisionFrame, faces : List[Face]) -> None:+ if numpy.any(vision_frame):+ vision_hash = create_hash(vision_frame.tobytes())+ FACE_STORE.setdefault(vision_hash,+ {+ 'lock': threading.Lock()+ })['faces'] = faces-def set_static_faces(vision_frame : VisionFrame, faces : List[Face]) -> None:- vision_hash = create_hash(vision_frame.tobytes())- if vision_hash:- FACE_STORE['static_faces'][vision_hash] = faces+def resolve_lock(vision_frame : VisionFrame) -> threading.Lock:+ if numpy.any(vision_frame):+ vision_hash = create_hash(vision_frame.tobytes())+ return FACE_STORE.setdefault(vision_hash,+ {+ 'lock': threading.Lock()+ }).get('lock')+ return threading.Lock()-def clear_static_faces() -> None:- FACE_STORE['static_faces'].clear()+def clear_faces() -> None:+ FACE_STORE.clear()diff --git a/facefusion/face_tracker.py b/facefusion/face_tracker.pynew file mode 100644index 00000000..c14afec3--- /dev/null+++ b/facefusion/face_tracker.py@@ -0,0 +1,61 @@+from typing import List++from facefusion.common_helper import get_first, get_last+from facefusion.face_creator import get_static_faces, refill_faces+from facefusion.face_helper import calculate_bounding_box_overlap+from facefusion.types import Face, FaceTrack, Score, VisionFrame+++def track_faces(vision_frames : List[VisionFrame], score : Score) -> List[Face]:+ target_index = len(vision_frames) // 2+ face_tracks = create_face_tracks(vision_frames, score)+ temp_faces = []++ for face_track in face_tracks:+ track_indices = sorted(face_track)+ track_index_first = get_first(track_indices)+ track_index_last = get_last(track_indices)+ track_range = range(track_index_first, track_index_last + 1)++ if target_index in track_range:+ fill_faces = []++ for index in track_range:+ fill_faces.append(face_track.get(index))++ temp_faces.append(refill_faces(fill_faces)[target_index - track_index_first])++ return temp_faces+++def create_face_tracks(vision_frames : List[VisionFrame], score : Score) -> List[FaceTrack]:+ face_tracks : List[FaceTrack] = []++ for frame_index, vision_frame in enumerate(vision_frames):+ for face in get_static_faces([ vision_frame ]):+ face_track = select_face_track(face_tracks, face, score)++ if face_track:+ face_track[frame_index] = face+ else:+ face_tracks.append(+ {+ frame_index : face+ })++ return face_tracks+++def select_face_track(face_tracks : List[FaceTrack], face : Face, score : Score) -> FaceTrack:+ select_track : FaceTrack = {}+ select_score = score++ for face_track in face_tracks:+ track_face = face_track.get(get_last(face_track))+ track_score = calculate_bounding_box_overlap(face.bounding_box, track_face.bounding_box)++ if track_score > select_score:+ select_score = track_score+ select_track = face_track++ return select_trackdiff --git a/facefusion/ffmpeg.py b/facefusion/ffmpeg.pyindex f26c77b7..a8ca32c1 100644--- a/facefusion/ffmpeg.py+++ b/facefusion/ffmpeg.py@@ -9,7 +9,7 @@ from tqdm import tqdmimport facefusion.choicesfrom facefusion import ffmpeg_builder, logger, process_manager, state_manager, translatorfrom facefusion.filesystem import get_file_format, remove_file-from facefusion.temp_helper import get_temp_file_path, get_temp_frames_pattern+from facefusion.temp_helper import get_temp_file_path, get_temp_frame_patternfrom facefusion.types import AudioBuffer, AudioEncoder, Command, EncoderSet, Fps, Resolution, UpdateProgress, VideoEncoder, VideoFormatfrom facefusion.vision import detect_video_duration, detect_video_fps, pack_resolution, predict_video_frame_total@@ -109,7 +109,7 @@ def get_available_encoder_set() -> EncoderSet:def extract_frames(target_path : str, temp_video_resolution : Resolution, temp_video_fps : Fps, trim_frame_start : int, trim_frame_end : int) -> bool:extract_frame_total = predict_video_frame_total(target_path, temp_video_fps, trim_frame_start, trim_frame_end)- temp_frames_pattern = get_temp_frames_pattern(target_path, '%08d')+ temp_frame_pattern = get_temp_frame_pattern(target_path, '%08d')commands = ffmpeg_builder.chain(ffmpeg_builder.set_input(target_path),ffmpeg_builder.set_media_resolution(pack_resolution(temp_video_resolution)),@@ -117,7 +117,8 @@ def extract_frames(target_path : str, temp_video_resolution : Resolution, temp_vffmpeg_builder.enforce_pixel_format('rgb24'),ffmpeg_builder.select_frame_range(trim_frame_start, trim_frame_end, temp_video_fps),ffmpeg_builder.prevent_frame_drop(),- ffmpeg_builder.set_output(temp_frames_pattern)+ ffmpeg_builder.set_start_number(trim_frame_start),+ ffmpeg_builder.set_output(temp_frame_pattern))with tqdm(total = extract_frame_total, desc = translator.get('extracting'), unit = 'frame', ascii = ' =', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:@@ -173,6 +174,7 @@ def restore_audio(target_path : str, output_path : str, trim_frame_start : int,temp_video_path = get_temp_file_path(target_path)temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))temp_video_duration = detect_video_duration(temp_video_path)+ output_video_format = cast(VideoFormat, get_file_format(output_path))output_audio_encoder = fix_audio_encoder(temp_video_format, output_audio_encoder)commands = ffmpeg_builder.chain(@@ -186,6 +188,7 @@ def restore_audio(target_path : str, output_path : str, trim_frame_start : int,ffmpeg_builder.select_media_stream('0:v:0'),ffmpeg_builder.select_media_stream('1:a:0'),ffmpeg_builder.set_video_duration(temp_video_duration),+ ffmpeg_builder.set_faststart(output_video_format),ffmpeg_builder.force_output(output_path))return run_ffmpeg(commands).returncode == 0@@ -198,6 +201,7 @@ def replace_audio(target_path : str, audio_path : str, output_path : str) -> bootemp_video_path = get_temp_file_path(target_path)temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))temp_video_duration = detect_video_duration(temp_video_path)+ output_video_format = cast(VideoFormat, get_file_format(output_path))output_audio_encoder = fix_audio_encoder(temp_video_format, output_audio_encoder)commands = ffmpeg_builder.chain(@@ -208,6 +212,7 @@ def replace_audio(target_path : str, audio_path : str, output_path : str) -> booffmpeg_builder.set_audio_quality(output_audio_encoder, output_audio_quality),ffmpeg_builder.set_audio_volume(output_audio_volume),ffmpeg_builder.set_video_duration(temp_video_duration),+ ffmpeg_builder.set_faststart(output_video_format),ffmpeg_builder.force_output(output_path))return run_ffmpeg(commands).returncode == 0@@ -220,14 +225,16 @@ def merge_video(target_path : str, temp_video_fps : Fps, output_video_resolutionmerge_frame_total = predict_video_frame_total(target_path, output_video_fps, trim_frame_start, trim_frame_end)temp_video_path = get_temp_file_path(target_path)temp_video_format = cast(VideoFormat, get_file_format(temp_video_path))- temp_frames_pattern = get_temp_frames_pattern(target_path, '%08d')+ temp_frame_pattern = get_temp_frame_pattern(target_path, '%08d')output_video_encoder = fix_video_encoder(temp_video_format, output_video_encoder)commands = ffmpeg_builder.chain(ffmpeg_builder.set_input_fps(temp_video_fps),- ffmpeg_builder.set_input(temp_frames_pattern),+ ffmpeg_builder.set_start_number(trim_frame_start),+ ffmpeg_builder.set_input(temp_frame_pattern),ffmpeg_builder.set_media_resolution(pack_resolution(output_video_resolution)),ffmpeg_builder.set_video_encoder(output_video_encoder),+ ffmpeg_builder.set_video_tag(output_video_encoder, temp_video_format),ffmpeg_builder.set_video_quality(output_video_encoder, output_video_quality),ffmpeg_builder.set_video_preset(output_video_encoder, output_video_preset),ffmpeg_builder.concat(@@ -254,11 +261,13 @@ def concat_video(output_path : str, temp_output_paths : List[str]) -> bool:concat_video_file.close()output_path = os.path.abspath(output_path)+ output_video_format = cast(VideoFormat, get_file_format(output_path))commands = ffmpeg_builder.chain(ffmpeg_builder.unsafe_concat(),ffmpeg_builder.set_input(concat_video_file.name),ffmpeg_builder.copy_video_encoder(),ffmpeg_builder.copy_audio_encoder(),+ ffmpeg_builder.set_faststart(output_video_format),ffmpeg_builder.force_output(output_path))process = run_ffmpeg(commands)diff --git a/facefusion/ffmpeg_builder.py b/facefusion/ffmpeg_builder.pyindex f62e5b26..60b6bd2c 100644--- a/facefusion/ffmpeg_builder.py+++ b/facefusion/ffmpeg_builder.py@@ -5,7 +5,7 @@ from typing import List, Optionalimport numpyfrom facefusion.filesystem import get_file_format-from facefusion.types import AudioEncoder, Command, CommandSet, Duration, Fps, StreamMode, VideoEncoder, VideoPreset+from facefusion.types import AudioEncoder, Command, CommandSet, Duration, Fps, StreamMode, VideoEncoder, VideoFormat, VideoPresetdef run(commands : List[Command]) -> List[Command]:@@ -51,6 +51,10 @@ def set_input_fps(input_fps : Fps) -> List[Command]:return [ '-r', str(input_fps) ]+def set_start_number(frame_number : int) -> List[Command]:+ return [ '-start_number', str(frame_number) ]++def set_output(output_path : str) -> List[Command]:return [ output_path ]@@ -187,6 +191,18 @@ def copy_video_encoder() -> List[Command]:return set_video_encoder('copy')+def set_faststart(video_format : VideoFormat) -> List[Command]:+ if video_format in [ 'm4v', 'mov', 'mp4' ]:+ return [ '-movflags', '+faststart' ]+ return []+++def set_video_tag(video_encoder : VideoEncoder, video_format : VideoFormat) -> List[Command]:+ if video_format in [ 'm4v', 'mov', 'mp4' ] and video_encoder in [ 'libx265', 'hevc_nvenc', 'hevc_amf', 'hevc_qsv', 'hevc_videotoolbox' ]:+ return [ '-tag:v', 'hvc1' ]+ return []++def set_video_quality(video_encoder : VideoEncoder, video_quality : int) -> List[Command]:if video_encoder in [ 'libx264', 'libx264rgb', 'libx265' ]:video_compression = numpy.round(numpy.interp(video_quality, [ 0, 100 ], [ 51, 0 ])).astype(int).item()diff --git a/facefusion/inference_manager.py b/facefusion/inference_manager.pyindex abcdeafa..3b10eed3 100644--- a/facefusion/inference_manager.py+++ b/facefusion/inference_manager.py@@ -1,5 +1,6 @@import importlibimport random+from functools import lru_cachefrom time import sleep, timefrom typing import List@@ -12,7 +13,7 @@ from facefusion.execution import create_inference_providers, has_execution_provifrom facefusion.exit_helper import fatal_exitfrom facefusion.filesystem import get_file_name, is_filefrom facefusion.time_helper import calculate_end_time-from facefusion.types import DownloadSet, ExecutionProvider, InferencePool, InferencePoolSet+from facefusion.types import DownloadSet, ExecutionProvider, InferencePool, InferencePoolSet, InferenceProviderINFERENCE_POOL_SET : InferencePoolSet =\{@@ -25,7 +26,7 @@ def get_inference_pool(module_name : str, model_names : List[str], model_source_while process_manager.is_checking():sleep(0.5)execution_device_ids = state_manager.get_item('execution_device_ids')- execution_providers = resolve_execution_providers(module_name)+ execution_providers = state_manager.get_item('execution_providers')app_context = detect_app_context()for execution_device_id in execution_device_ids:@@ -36,26 +37,27 @@ def get_inference_pool(module_name : str, model_names : List[str], model_source_if app_context == 'ui' and INFERENCE_POOL_SET.get('cli').get(inference_context):INFERENCE_POOL_SET['ui'][inference_context] = INFERENCE_POOL_SET.get('cli').get(inference_context)if not INFERENCE_POOL_SET.get(app_context).get(inference_context):- INFERENCE_POOL_SET[app_context][inference_context] = create_inference_pool(model_source_set, execution_device_id, execution_providers)+ inference_providers = resolve_static_inference_providers(module_name, execution_device_id)+ INFERENCE_POOL_SET[app_context][inference_context] = create_inference_pool(model_source_set, inference_providers)current_inference_context = get_inference_context(module_name, model_names, random.choice(execution_device_ids), execution_providers)return INFERENCE_POOL_SET.get(app_context).get(current_inference_context)-def create_inference_pool(model_source_set : DownloadSet, execution_device_id : int, execution_providers : List[ExecutionProvider]) -> InferencePool:+def create_inference_pool(model_source_set : DownloadSet, inference_providers : List[InferenceProvider]) -> InferencePool:inference_pool : InferencePool = {}for model_name in model_source_set.keys():model_path = model_source_set.get(model_name).get('path')if is_file(model_path):- inference_pool[model_name] = create_inference_session(model_path, execution_device_id, execution_providers)+ inference_pool[model_name] = create_inference_session(model_path, inference_providers)return inference_pooldef clear_inference_pool(module_name : str, model_names : List[str]) -> None:execution_device_ids = state_manager.get_item('execution_device_ids')- execution_providers = resolve_execution_providers(module_name)+ execution_providers = state_manager.get_item('execution_providers')app_context = detect_app_context()if is_windows() and has_execution_provider('directml'):@@ -67,12 +69,11 @@ def clear_inference_pool(module_name : str, model_names : List[str]) -> None:del INFERENCE_POOL_SET[app_context][inference_context]-def create_inference_session(model_path : str, execution_device_id : int, execution_providers : List[ExecutionProvider]) -> InferenceSession:+def create_inference_session(model_path : str, inference_providers : List[InferenceProvider]) -> InferenceSession:model_file_name = get_file_name(model_path)start_time = time()try:- inference_providers = create_inference_providers(execution_device_id, execution_providers)inference_session = InferenceSession(model_path, providers = inference_providers)logger.debug(translator.get('loading_model_succeeded').format(model_name = model_file_name, seconds = calculate_end_time(start_time)), __name__)return inference_session@@ -87,9 +88,15 @@ def get_inference_context(module_name : str, model_names : List[str], execution_return inference_context-def resolve_execution_providers(module_name : str) -> List[ExecutionProvider]:+@lru_cache()+def resolve_static_inference_providers(module_name : str, execution_device_id : int) -> List[InferenceProvider]:module = importlib.import_module(module_name)+ execution_providers = state_manager.get_item('execution_providers')- if hasattr(module, 'resolve_execution_providers'):- return getattr(module, 'resolve_execution_providers')()- return state_manager.get_item('execution_providers')+ if hasattr(module, 'resolve_inference_providers'):+ inference_providers = getattr(module, 'resolve_inference_providers')()++ if inference_providers:+ return inference_providers++ return create_inference_providers(execution_device_id, execution_providers)diff --git a/facefusion/installer.py b/facefusion/installer.pyindex f50a0c37..7bbcaa7c 100644--- a/facefusion/installer.py+++ b/facefusion/installer.py@@ -19,23 +19,23 @@ LOCALES =\}ONNXRUNTIME_SET =\{- 'default': ('onnxruntime', '1.24.4')+ 'default': ('onnxruntime', '1.26.0')}if is_windows() or is_linux():- ONNXRUNTIME_SET['cuda'] = ('onnxruntime-gpu', '1.24.4')+ ONNXRUNTIME_SET['cuda'] = ('onnxruntime-gpu', '1.26.0')ONNXRUNTIME_SET['openvino'] = ('onnxruntime-openvino', '1.24.1')if is_windows():ONNXRUNTIME_SET['directml'] = ('onnxruntime-directml', '1.24.4')ONNXRUNTIME_SET['qnn'] = ('onnxruntime-qnn', '1.24.4')if is_linux():- ONNXRUNTIME_SET['migraphx'] = ('onnxruntime-migraphx', '1.24.2')+ ONNXRUNTIME_SET['migraphx'] = ('onnxruntime-migraphx', '1.25.0')ONNXRUNTIME_SET['rocm'] = ('onnxruntime-rocm', '1.22.2.post1')def cli() -> None:signal.signal(signal.SIGINT, signal_exit)program = ArgumentParser(formatter_class = partial(HelpFormatter, max_help_position = 50))- program.add_argument('--onnxruntime', help = LOCALES.get('install_dependency').format(dependency = 'onnxruntime'), choices = ONNXRUNTIME_SET.keys(), required = True)+ program.add_argument('onnxruntime', help = LOCALES.get('install_dependency').format(dependency = 'onnxruntime'), choices = ONNXRUNTIME_SET.keys())program.add_argument('--force-reinstall', help = LOCALES.get('force_reinstall'), action = 'store_true')program.add_argument('--skip-conda', help = LOCALES.get('skip_conda'), action = 'store_true')program.add_argument('-v', '--version', version = metadata.get('name') + ' ' + metadata.get('version'), action = 'version')diff --git a/facefusion/jobs/job_manager.py b/facefusion/jobs/job_manager.pyindex 8abc34b8..4c9acda8 100644--- a/facefusion/jobs/job_manager.py+++ b/facefusion/jobs/job_manager.py@@ -6,6 +6,7 @@ import facefusion.choicesfrom facefusion.filesystem import create_directory, get_file_name, is_directory, is_file, move_file, remove_directory, remove_file, resolve_file_patternfrom facefusion.jobs.job_helper import get_step_output_pathfrom facefusion.json import read_json, write_json+from facefusion.sanitizer import sanitize_job_idfrom facefusion.time_helper import get_current_date_timefrom facefusion.types import Args, Job, JobSet, JobStatus, JobStep, JobStepStatus@@ -261,5 +262,6 @@ def find_job_path(job_id : str) -> Optional[str]:def get_job_file_name(job_id : str) -> Optional[str]:if job_id:+ job_id = sanitize_job_id(job_id)return job_id + '.json'return Nonediff --git a/facefusion/locales.py b/facefusion/locales.pyindex 7d0f8ad4..3fc95751 100644--- a/facefusion/locales.py+++ b/facefusion/locales.py@@ -124,6 +124,7 @@ LOCALES : Locales =\'reference_face_position': 'specify the position used to create the reference face','reference_face_distance': 'specify the similarity between the reference face and target face','reference_frame_number': 'specify the frame used to create the reference face',+ 'face_tracker_score': 'specify the minimum score to track a face','face_occluder_model': 'choose the model responsible for the occlusion mask','face_parser_model': 'choose the model responsible for the region mask','face_mask_types': 'mix and match different face mask types (choices: {choices})',@@ -136,6 +137,7 @@ LOCALES : Locales =\'trim_frame_end': 'specify the ending frame of the target video','temp_frame_format': 'specify the temporary resources format','keep_temp': 'keep the temporary resources after processing',+ 'target_frame_amount': 'specify the amount of frames around the target frame','output_image_quality': 'specify the image quality which translates to the image compression','output_image_scale': 'specify the image scale based on the target image','output_audio_encoder': 'specify the encoder used for the audio',@@ -161,7 +163,6 @@ LOCALES : Locales =\'execution_providers': 'inference using different providers (choices: {choices}, ...)','execution_thread_count': 'specify the amount of parallel threads while processing','video_memory_strategy': 'balance fast processing and low VRAM usage',- 'system_memory_limit': 'limit the available RAM that can be used while processing','log_level': 'adjust the message severity displayed in the terminal','halt_on_error': 'halt the program once an error occurred','run': 'run the program',@@ -224,6 +225,7 @@ LOCALES : Locales =\'face_selector_mode_dropdown': 'FACE SELECTOR MODE','face_selector_order_dropdown': 'FACE SELECTOR ORDER','face_selector_race_dropdown': 'FACE SELECTOR RACE',+ 'face_tracker_score_slider': 'FACE TRACKER SCORE','face_occluder_model_dropdown': 'FACE OCCLUDER MODEL','face_parser_model_dropdown': 'FACE PARSER MODEL','voice_extractor_model_dropdown': 'VOICE EXTRACTOR MODEL',@@ -257,7 +259,6 @@ LOCALES : Locales =\'source_file': 'SOURCE','start_button': 'START','stop_button': 'STOP',- 'system_memory_limit_slider': 'SYSTEM MEMORY LIMIT','target_file': 'TARGET','temp_frame_format_dropdown': 'TEMP FRAME FORMAT','terminal_textbox': 'TERMINAL',diff --git a/facefusion/memory.py b/facefusion/memory.pydeleted file mode 100644index f4161ac0..00000000--- a/facefusion/memory.py+++ /dev/null@@ -1,21 +0,0 @@-from facefusion.common_helper import is_macos, is_windows--if is_windows():- import ctypes-else:- import resource---def limit_system_memory(system_memory_limit : int = 1) -> bool:- if is_macos():- system_memory_limit = system_memory_limit * (1024 ** 6)- else:- system_memory_limit = system_memory_limit * (1024 ** 3)- try:- if is_windows():- ctypes.windll.kernel32.SetProcessWorkingSetSize(-1, ctypes.c_size_t(system_memory_limit), ctypes.c_size_t(system_memory_limit)) #type:ignore[attr-defined]- else:- resource.setrlimit(resource.RLIMIT_DATA, (system_memory_limit, system_memory_limit))- return True- except Exception:- return Falsediff --git a/facefusion/metadata.py b/facefusion/metadata.pyindex 027d5aed..4cc03d69 100644--- a/facefusion/metadata.py+++ b/facefusion/metadata.py@@ -4,7 +4,7 @@ METADATA =\{'name': 'FaceFusion','description': 'Industry leading face manipulation platform',- 'version': '3.6.1',+ 'version': '3.7.0','license': 'OpenRAIL-AS','author': 'Henry Ruhs','url': 'https://facefusion.io'diff --git a/facefusion/processors/core.py b/facefusion/processors/core.pyindex 09d45e5e..fe395445 100644--- a/facefusion/processors/core.py+++ b/facefusion/processors/core.py@@ -12,6 +12,7 @@ PROCESSORS_METHODS =\'clear_inference_pool','register_args','apply_args',+ 'get_common_modules','pre_check','pre_process','post_process',diff --git a/facefusion/processors/modules/age_modifier/core.py b/facefusion/processors/modules/age_modifier/core.pyindex 385c2a56..ceb6832d 100755--- a/facefusion/processors/modules/age_modifier/core.py+++ b/facefusion/processors/modules/age_modifier/core.py@@ -1,5 +1,7 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleType+from typing import Listimport cv2import numpy@@ -8,10 +10,9 @@ import facefusion.choicesimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import create_int_metavar, is_macos+from facefusion.common_helper import create_int_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.execution import has_execution_provider-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import merge_matrix, paste_back, scale_face_landmark_5, warp_face_by_face_landmark_5from facefusion.face_masker import create_box_mask, create_occlusion_maskfrom facefusion.face_selector import select_faces@@ -22,7 +23,7 @@ from facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import thread_semaphorefrom facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import match_frame_color, read_static_image, read_static_video_frame+from facefusion.vision import match_frame_color, read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -134,10 +135,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('age_modifier_direction', args.get('age_modifier_direction'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -157,16 +166,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def modify_age(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -231,9 +239,6 @@ def forward(crop_vision_frame : VisionFrame, extend_vision_frame : VisionFrame,age_modifier = get_inference_pool().get('age_modifier')age_modifier_inputs = {}- if is_macos() and has_execution_provider('coreml'):- age_modifier.set_providers([ facefusion.choices.execution_provider_set.get('cpu') ])-for age_modifier_input in age_modifier.get_inputs():if age_modifier_input.name == 'target':age_modifier_inputs[age_modifier_input.name] = crop_vision_frame@@ -280,10 +285,13 @@ def normalize_extend_frame(extend_vision_frame : VisionFrame) -> VisionFrame:def process_frame(inputs : AgeModifierInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/age_modifier/types.py b/facefusion/processors/modules/age_modifier/types.pyindex 3a204abc..5f968131 100644--- a/facefusion/processors/modules/age_modifier/types.py+++ b/facefusion/processors/modules/age_modifier/types.py@@ -1,4 +1,4 @@-from typing import Any, Literal, TypeAlias, TypedDict+from typing import Any, List, Literal, TypeAlias, TypedDictfrom numpy.typing import NDArray@@ -7,7 +7,8 @@ from facefusion.types import Mask, VisionFrameAgeModifierInputs = TypedDict('AgeModifierInputs',{'reference_vision_frame' : VisionFrame,- 'target_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/background_remover/core.py b/facefusion/processors/modules/background_remover/core.pyindex e1dbe82c..11f3ca26 100644--- a/facefusion/processors/modules/background_remover/core.py+++ b/facefusion/processors/modules/background_remover/core.py@@ -1,10 +1,12 @@from argparse import ArgumentParserfrom functools import lru_cache, partial+from types import ModuleTypefrom typing import List, Tupleimport cv2import numpy+import facefusion.choicesimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager@@ -19,8 +21,8 @@ from facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.sanitizer import sanitize_int_rangefrom facefusion.thread_helper import thread_semaphore-from facefusion.types import ApplyStateItem, Args, DownloadScope, ExecutionProvider, InferencePool, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_video_frame+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -477,12 +479,13 @@ def clear_inference_pool() -> None:inference_manager.clear_inference_pool(__name__, model_names)-def resolve_execution_providers() -> List[ExecutionProvider]:+def resolve_inference_providers() -> List[InferenceProvider]:model_type = get_model_options().get('type')if is_macos() and has_execution_provider('coreml') or is_windows() and has_execution_provider('directml') and model_type == 'corridor_key':- return [ 'cpu' ]- return state_manager.get_item('execution_providers')+ return [ facefusion.choices.execution_provider_set.get('cpu') ]++ return []def get_model_options() -> ModelOptions:@@ -505,10 +508,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('background_remover_despill_color', normalize_color(args.get('background_remover_despill_color')))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -528,11 +539,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def remove_background(temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, Mask]:diff --git a/facefusion/processors/modules/background_remover/types.py b/facefusion/processors/modules/background_remover/types.pyindex e2c5290f..6e19dc0f 100644--- a/facefusion/processors/modules/background_remover/types.py+++ b/facefusion/processors/modules/background_remover/types.py@@ -1,10 +1,10 @@-from typing import Literal, TypedDict+from typing import List, Literal, TypedDictfrom facefusion.types import Mask, VisionFrameBackgroundRemoverInputs = TypedDict('BackgroundRemoverInputs',{- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/deep_swapper/core.py b/facefusion/processors/modules/deep_swapper/core.pyindex dcb9bb2b..8c3661dc 100755--- a/facefusion/processors/modules/deep_swapper/core.py+++ b/facefusion/processors/modules/deep_swapper/core.py@@ -1,6 +1,7 @@from argparse import ArgumentParserfrom functools import lru_cache-from typing import Tuple+from types import ModuleType+from typing import List, Tupleimport cv2import numpy@@ -9,9 +10,9 @@ from cv2.typing import Sizeimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import create_int_metavar+from facefusion.common_helper import create_int_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url_by_provider-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import paste_back, warp_face_by_face_landmark_5from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_maskfrom facefusion.face_selector import select_faces@@ -22,7 +23,7 @@ from facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import thread_semaphorefrom facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, Mask, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import conditional_match_frame_color, read_static_image, read_static_video_frame+from facefusion.vision import conditional_match_frame_color, read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -286,10 +287,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('deep_swapper_morph', args.get('deep_swapper_morph'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+if model_hash_set and model_source_set:return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)return True@@ -311,16 +320,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def swap_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -411,10 +419,13 @@ def prepare_crop_mask(crop_source_mask : Mask, crop_target_mask : Mask) -> Mask:def process_frame(inputs : DeepSwapperInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/deep_swapper/types.py b/facefusion/processors/modules/deep_swapper/types.pyindex b404fef8..41e4f979 100644--- a/facefusion/processors/modules/deep_swapper/types.py+++ b/facefusion/processors/modules/deep_swapper/types.py@@ -1,4 +1,4 @@-from typing import Any, TypeAlias, TypedDict+from typing import Any, List, TypeAlias, TypedDictfrom numpy.typing import NDArray@@ -7,7 +7,8 @@ from facefusion.types import Mask, VisionFrameDeepSwapperInputs = TypedDict('DeepSwapperInputs',{'reference_vision_frame' : VisionFrame,- 'target_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/expression_restorer/core.py b/facefusion/processors/modules/expression_restorer/core.pyindex c6c4f963..ed94fd91 100755--- a/facefusion/processors/modules/expression_restorer/core.py+++ b/facefusion/processors/modules/expression_restorer/core.py@@ -1,6 +1,7 @@from argparse import ArgumentParserfrom functools import lru_cache-from typing import Tuple+from types import ModuleType+from typing import List, Tupleimport cv2import numpy@@ -8,9 +9,9 @@ import numpyimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import create_int_metavar+from facefusion.common_helper import create_int_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import paste_back, warp_face_by_face_landmark_5from facefusion.face_masker import create_box_mask, create_occlusion_maskfrom facefusion.face_selector import select_faces@@ -22,7 +23,7 @@ from facefusion.processors.types import LivePortraitExpression, LivePortraitFeatfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import conditional_thread_semaphore, thread_semaphorefrom facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_video_frame+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -111,10 +112,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('expression_restorer_areas', args.get('expression_restorer_areas'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -137,16 +146,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def restore_expression(target_face : Face, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -257,10 +265,13 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:def process_frame(inputs : ExpressionRestorerInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/expression_restorer/types.py b/facefusion/processors/modules/expression_restorer/types.pyindex 1cee5a52..e0dec6e3 100644--- a/facefusion/processors/modules/expression_restorer/types.py+++ b/facefusion/processors/modules/expression_restorer/types.py@@ -6,7 +6,7 @@ ExpressionRestorerInputs = TypedDict('ExpressionRestorerInputs',{'reference_vision_frame' : VisionFrame,'source_vision_frames' : List[VisionFrame],- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/face_debugger/core.py b/facefusion/processors/modules/face_debugger/core.pyindex ab649d93..6ff15310 100755--- a/facefusion/processors/modules/face_debugger/core.py+++ b/facefusion/processors/modules/face_debugger/core.py@@ -1,4 +1,6 @@from argparse import ArgumentParser+from types import ModuleType+from typing import Listimport cv2import numpy@@ -6,7 +8,8 @@ import numpyimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, logger, state_manager, translator, video_manager-from facefusion.face_analyser import scale_face+from facefusion.common_helper import get_middle+from facefusion.face_creator import scale_facefrom facefusion.face_helper import warp_face_by_face_landmark_5from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_maskfrom facefusion.face_selector import select_faces@@ -16,7 +19,7 @@ from facefusion.processors.modules.face_debugger.types import FaceDebuggerInputsfrom facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.types import ApplyStateItem, Args, Face, InferencePool, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_video_frame+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_framedef get_inference_pool() -> InferencePool:@@ -38,7 +41,14 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('face_debugger_items', args.get('face_debugger_items'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return Falsereturn True@@ -58,14 +68,12 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def debug_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -94,21 +102,22 @@ def debug_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFradef draw_bounding_box(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)- box_color = 0, 0, 255- border_color = 100, 100, 255bounding_box = target_face.bounding_box.astype(numpy.int32)x1, y1, x2, y2 = bounding_box+ box_color = 0, 0, 255+ border_scale = calculate_scale(temp_vision_frame)+ border_color = 100, 100, 255- cv2.rectangle(temp_vision_frame, (x1, y1), (x2, y2), box_color, 2)+ cv2.rectangle(temp_vision_frame, (x1, y1), (x2, y2), box_color, border_scale)if target_face.angle == 0:- cv2.line(temp_vision_frame, (x1, y1), (x2, y1), border_color, 3)+ cv2.line(temp_vision_frame, (x1, y1), (x2, y1), border_color, border_scale + 1)if target_face.angle == 180:- cv2.line(temp_vision_frame, (x1, y2), (x2, y2), border_color, 3)+ cv2.line(temp_vision_frame, (x1, y2), (x2, y2), border_color, border_scale + 1)if target_face.angle == 90:- cv2.line(temp_vision_frame, (x2, y1), (x2, y2), border_color, 3)+ cv2.line(temp_vision_frame, (x2, y1), (x2, y2), border_color, border_scale + 1)if target_face.angle == 270:- cv2.line(temp_vision_frame, (x1, y1), (x1, y2), border_color, 3)+ cv2.line(temp_vision_frame, (x1, y1), (x1, y2), border_color, border_scale + 1)return temp_vision_frame@@ -122,11 +131,15 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visiocrop_vision_frame, affine_matrix = warp_face_by_face_landmark_5(temp_vision_frame, face_landmark_5_68, 'arcface_128', (512, 512))inverse_matrix = cv2.invertAffineTransform(affine_matrix)temp_size = temp_vision_frame.shape[:2][::-1]+ mask_scale = calculate_scale(temp_vision_frame)mask_color = 0, 255, 0if numpy.array_equal(face_landmark_5, face_landmark_5_68):mask_color = 255, 255, 0+ if target_face.origin == 'refill':+ mask_color = 0, 165, 255+if 'box' in state_manager.get_item('face_mask_types'):box_mask = create_box_mask(crop_vision_frame, 0, state_manager.get_item('face_mask_padding'))crop_masks.append(box_mask)@@ -149,7 +162,7 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visioinverse_vision_frame = cv2.warpAffine(crop_mask, inverse_matrix, temp_size)inverse_vision_frame = cv2.threshold(inverse_vision_frame, 100, 255, cv2.THRESH_BINARY)[1]inverse_contours, _ = cv2.findContours(inverse_vision_frame, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)- cv2.drawContours(temp_vision_frame, inverse_contours, -1, mask_color, 2)+ cv2.drawContours(temp_vision_frame, inverse_contours, -1, mask_color, mask_scale)return temp_vision_frame@@ -157,13 +170,17 @@ def draw_face_mask(target_face : Face, temp_vision_frame : VisionFrame) -> Visiodef draw_face_landmark_5(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)face_landmark_5 = target_face.landmark_set.get('5')+ point_scale = calculate_scale(temp_vision_frame)point_color = 0, 0, 255+ if target_face.origin == 'refill':+ point_color = 0, 165, 255+if numpy.any(face_landmark_5):face_landmark_5 = face_landmark_5.astype(numpy.int32)for point in face_landmark_5:- cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)+ cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)return temp_vision_frame@@ -172,16 +189,20 @@ def draw_face_landmark_5_68(target_face : Face, temp_vision_frame : VisionFrame)temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)face_landmark_5 = target_face.landmark_set.get('5')face_landmark_5_68 = target_face.landmark_set.get('5/68')+ point_scale = calculate_scale(temp_vision_frame)point_color = 0, 255, 0if numpy.array_equal(face_landmark_5, face_landmark_5_68):point_color = 255, 255, 0+ if target_face.origin == 'refill':+ point_color = 0, 165, 255+if numpy.any(face_landmark_5_68):face_landmark_5_68 = face_landmark_5_68.astype(numpy.int32)for point in face_landmark_5_68:- cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)+ cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)return temp_vision_frame@@ -190,16 +211,20 @@ def draw_face_landmark_68(target_face : Face, temp_vision_frame : VisionFrame) -temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)face_landmark_68 = target_face.landmark_set.get('68')face_landmark_68_5 = target_face.landmark_set.get('68/5')+ point_scale = calculate_scale(temp_vision_frame)point_color = 0, 255, 0if numpy.array_equal(face_landmark_68, face_landmark_68_5):point_color = 255, 255, 0+ if target_face.origin == 'refill':+ point_color = 0, 165, 255+if numpy.any(face_landmark_68):face_landmark_68 = face_landmark_68.astype(numpy.int32)for point in face_landmark_68:- cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)+ cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)return temp_vision_frame@@ -207,23 +232,36 @@ def draw_face_landmark_68(target_face : Face, temp_vision_frame : VisionFrame) -def draw_face_landmark_68_5(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:temp_vision_frame = numpy.ascontiguousarray(temp_vision_frame)face_landmark_68_5 = target_face.landmark_set.get('68/5')+ point_scale = calculate_scale(temp_vision_frame)point_color = 255, 255, 0+ if target_face.origin == 'refill':+ point_color = 0, 165, 255+if numpy.any(face_landmark_68_5):face_landmark_68_5 = face_landmark_68_5.astype(numpy.int32)for point in face_landmark_68_5:- cv2.circle(temp_vision_frame, tuple(point), 3, point_color, -1)+ cv2.circle(temp_vision_frame, tuple(point), point_scale, point_color, -1)return temp_vision_frame+def calculate_scale(temp_vision_frame : VisionFrame) -> int:+ frame_height, _ = temp_vision_frame.shape[:2]+ frame_scale = round(frame_height / 270)+ return max(1, min(10, frame_scale))++def process_frame(inputs : FaceDebuggerInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:@@ -231,5 +269,3 @@ def process_frame(inputs : FaceDebuggerInputs) -> ProcessorOutputs:temp_vision_frame = debug_face(target_face, temp_vision_frame)return temp_vision_frame, temp_vision_mask--diff --git a/facefusion/processors/modules/face_debugger/types.py b/facefusion/processors/modules/face_debugger/types.pyindex a51539b6..afb7e5b3 100644--- a/facefusion/processors/modules/face_debugger/types.py+++ b/facefusion/processors/modules/face_debugger/types.py@@ -1,11 +1,12 @@-from typing import Literal, TypedDict+from typing import List, Literal, TypedDictfrom facefusion.types import Mask, VisionFrameFaceDebuggerInputs = TypedDict('FaceDebuggerInputs',{'reference_vision_frame' : VisionFrame,- 'target_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/face_editor/core.py b/facefusion/processors/modules/face_editor/core.pyindex cd5ed964..78e1e303 100755--- a/facefusion/processors/modules/face_editor/core.py+++ b/facefusion/processors/modules/face_editor/core.py@@ -1,6 +1,7 @@from argparse import ArgumentParserfrom functools import lru_cache-from typing import Tuple+from types import ModuleType+from typing import List, Tupleimport cv2import numpy@@ -8,9 +9,9 @@ import numpyimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import create_float_metavar+from facefusion.common_helper import create_float_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import paste_back, scale_face_landmark_5, warp_face_by_face_landmark_5from facefusion.face_masker import create_box_maskfrom facefusion.face_selector import select_faces@@ -22,7 +23,7 @@ from facefusion.processors.types import LivePortraitExpression, LivePortraitFeatfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import conditional_thread_semaphore, thread_semaphorefrom facefusion.types import ApplyStateItem, Args, DownloadScope, Face, FaceLandmark68, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_video_frame+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -165,10 +166,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('face_editor_head_roll', args.get('face_editor_head_roll'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -188,16 +197,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def edit_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -486,10 +494,13 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:def process_frame(inputs : FaceEditorInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/face_editor/types.py b/facefusion/processors/modules/face_editor/types.pyindex 6e246466..2df890ff 100644--- a/facefusion/processors/modules/face_editor/types.py+++ b/facefusion/processors/modules/face_editor/types.py@@ -1,11 +1,12 @@-from typing import Literal, TypedDict+from typing import List, Literal, TypedDictfrom facefusion.types import Mask, VisionFrameFaceEditorInputs = TypedDict('FaceEditorInputs',{'reference_vision_frame' : VisionFrame,- 'target_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/face_enhancer/core.py b/facefusion/processors/modules/face_enhancer/core.pyindex 9a1854b0..b81cfe67 100755--- a/facefusion/processors/modules/face_enhancer/core.py+++ b/facefusion/processors/modules/face_enhancer/core.py@@ -1,14 +1,16 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleType+from typing import Listimport numpyimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import create_float_metavar, create_int_metavar+from facefusion.common_helper import create_float_metavar, create_int_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import paste_back, warp_face_by_face_landmark_5from facefusion.face_masker import create_box_mask, create_occlusion_maskfrom facefusion.face_selector import select_faces@@ -19,7 +21,7 @@ from facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import thread_semaphorefrom facefusion.types import ApplyStateItem, Args, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import blend_frame, read_static_image, read_static_video_frame+from facefusion.vision import blend_frame, read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -304,10 +306,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('face_enhancer_weight', args.get('face_enhancer_weight'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -327,16 +337,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def enhance_face(target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -413,10 +422,13 @@ def blend_paste_frame(temp_vision_frame : VisionFrame, paste_vision_frame : Visidef process_frame(inputs : FaceEnhancerInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')- target_vision_frame = inputs.get('target_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/face_enhancer/types.py b/facefusion/processors/modules/face_enhancer/types.pyindex 484a98c6..2104dbb8 100644--- a/facefusion/processors/modules/face_enhancer/types.py+++ b/facefusion/processors/modules/face_enhancer/types.py@@ -1,4 +1,4 @@-from typing import Any, Literal, TypeAlias, TypedDict+from typing import Any, List, Literal, TypeAlias, TypedDictfrom numpy.typing import NDArray@@ -7,7 +7,8 @@ from facefusion.types import Mask, VisionFrameFaceEnhancerInputs = TypedDict('FaceEnhancerInputs',{'reference_vision_frame' : VisionFrame,- 'target_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/face_swapper/core.py b/facefusion/processors/modules/face_swapper/core.pyindex 2ab17194..058c12aa 100755--- a/facefusion/processors/modules/face_swapper/core.py+++ b/facefusion/processors/modules/face_swapper/core.py@@ -1,5 +1,6 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleTypefrom typing import List, Optional, Tupleimport cv2@@ -9,10 +10,10 @@ import facefusion.choicesimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager-from facefusion.common_helper import get_first, is_macos+from facefusion.common_helper import get_first, get_middle, is_macosfrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_urlfrom facefusion.execution import has_execution_provider-from facefusion.face_analyser import get_average_face, get_many_faces, get_one_face, scale_face+from facefusion.face_creator import average_face_identity, get_one_face, get_static_faces, scale_facefrom facefusion.face_helper import paste_back, warp_face_by_face_landmark_5from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_mask, create_region_maskfrom facefusion.face_selector import select_faces, sort_faces_by_order@@ -24,8 +25,8 @@ from facefusion.processors.pixel_boost import explode_pixel_boost, implode_pixelfrom facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import conditional_thread_semaphore-from facefusion.types import ApplyStateItem, Args, DownloadScope, Embedding, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_images, read_static_video_frame, unpack_resolution+from facefusion.types import ApplyStateItem, Args, DownloadScope, Embedding, Face, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame+from facefusion.vision import read_static_image, read_static_images, read_static_video_chunk, read_static_video_frame, unpack_resolution@lru_cache()@@ -246,6 +247,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/hyperswap_1a_256.onnx')}},+ 'precision': 'fp16','type': 'hyperswap','template': 'arcface_128','size': (256, 256),@@ -276,6 +278,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/hyperswap_1b_256.onnx')}},+ 'precision': 'fp16','type': 'hyperswap','template': 'arcface_128','size': (256, 256),@@ -306,6 +309,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/hyperswap_1c_256.onnx')}},+ 'precision': 'fp16','type': 'hyperswap','template': 'arcface_128','size': (256, 256),@@ -366,6 +370,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/inswapper_128_fp16.onnx')}},+ 'precision': 'fp16','type': 'inswapper','template': 'arcface_128','size': (128, 128),@@ -486,28 +491,38 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:def get_inference_pool() -> InferencePool:- model_names = [ get_model_name() ]+ model_names = [ state_manager.get_item('face_swapper_model') ]model_source_set = get_model_options().get('sources')return inference_manager.get_inference_pool(__name__, model_names, model_source_set)def clear_inference_pool() -> None:- model_names = [ get_model_name() ]+ model_names = [ state_manager.get_item('face_swapper_model') ]inference_manager.clear_inference_pool(__name__, model_names)-def get_model_options() -> ModelOptions:- model_name = get_model_name()- return create_static_model_set('full').get(model_name)+def resolve_inference_providers() -> List[InferenceProvider]:+ model_precision = get_model_options().get('precision')+ model_type = get_model_options().get('type')+ if is_macos() and has_execution_provider('coreml'):+ if model_type in [ 'ghost', 'uniface' ] or model_precision == 'fp16':+ return\+ [+ (facefusion.choices.execution_provider_set.get('coreml'),+ {+ 'ModelFormat': 'MLProgram',+ 'SpecializationStrategy': 'FastPrediction'+ })+ ]-def get_model_name() -> str:- model_name = state_manager.get_item('face_swapper_model')+ return []- if is_macos() and has_execution_provider('coreml') and model_name == 'inswapper_128_fp16':- return 'inswapper_128'- return model_name++def get_model_options() -> ModelOptions:+ model_name = state_manager.get_item('face_swapper_model')+ return create_static_model_set('full').get(model_name)def register_args(program : ArgumentParser) -> None:@@ -527,10 +542,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('face_swapper_weight', args.get('face_swapper_weight'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -541,7 +564,7 @@ def pre_process(mode : ProcessMode) -> bool:source_image_paths = filter_image_paths(state_manager.get_item('source_paths'))source_vision_frames = read_static_images(source_image_paths)- source_faces = get_many_faces(source_vision_frames)+ source_faces = get_static_faces(source_vision_frames)if not get_one_face(source_faces):logger.error(translator.get('no_source_face_detected') + translator.get('exclamation_mark'), __name__)@@ -565,20 +588,19 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:get_static_model_initializer.cache_clear()clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()-def swap_face(source_face : Face, target_face : Face, temp_vision_frame : VisionFrame) -> VisionFrame:+def swap_face(source_face : Face, target_face : Face, source_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> VisionFrame:model_template = get_model_options().get('template')model_size = get_model_options().get('size')pixel_boost_size = unpack_resolution(state_manager.get_item('face_swapper_pixel_boost'))@@ -598,7 +620,7 @@ def swap_face(source_face : Face, target_face : Face, temp_vision_frame : Visionpixel_boost_vision_frames = implode_pixel_boost(crop_vision_frame, pixel_boost_total, model_size)for pixel_boost_vision_frame in pixel_boost_vision_frames:pixel_boost_vision_frame = prepare_crop_frame(pixel_boost_vision_frame)- pixel_boost_vision_frame = forward_swap_face(source_face, target_face, pixel_boost_vision_frame)+ pixel_boost_vision_frame = forward_swap_face(source_face, target_face, source_vision_frame, pixel_boost_vision_frame)pixel_boost_vision_frame = normalize_crop_frame(pixel_boost_vision_frame)temp_vision_frames.append(pixel_boost_vision_frame)crop_vision_frame = explode_pixel_boost(temp_vision_frames, pixel_boost_total, model_size, pixel_boost_size)@@ -617,18 +639,15 @@ def swap_face(source_face : Face, target_face : Face, temp_vision_frame : Visionreturn paste_vision_frame-def forward_swap_face(source_face : Face, target_face : Face, crop_vision_frame : VisionFrame) -> VisionFrame:+def forward_swap_face(source_face : Face, target_face : Face, source_vision_frame : VisionFrame, crop_vision_frame : VisionFrame) -> VisionFrame:face_swapper = get_inference_pool().get('face_swapper')model_type = get_model_options().get('type')face_swapper_inputs = {}- if is_macos() and has_execution_provider('coreml') and model_type in [ 'ghost', 'uniface' ]:- face_swapper.set_providers([ facefusion.choices.execution_provider_set.get('cpu') ])-for face_swapper_input in face_swapper.get_inputs():if face_swapper_input.name == 'source':if model_type in [ 'blendswap', 'uniface' ]:- face_swapper_inputs[face_swapper_input.name] = prepare_source_frame(source_face)+ face_swapper_inputs[face_swapper_input.name] = prepare_source_frame(source_face, source_vision_frame)else:source_embedding = prepare_source_embedding(source_face)source_embedding = balance_source_embedding(source_embedding, target_face.embedding)@@ -654,9 +673,8 @@ def forward_convert_embedding(face_embedding : Embedding) -> Embedding:return face_embedding-def prepare_source_frame(source_face : Face) -> VisionFrame:+def prepare_source_frame(source_face : Face, source_vision_frame : VisionFrame) -> VisionFrame:model_type = get_model_options().get('type')- source_vision_frame = read_static_image(get_first(state_manager.get_item('source_paths')))if model_type == 'blendswap':source_vision_frame, _ = warp_face_by_face_landmark_5(source_vision_frame, source_face.landmark_set.get('5/68'), 'arcface_112_v2', (112, 112))@@ -748,27 +766,31 @@ def extract_source_face(source_vision_frames : List[VisionFrame]) -> Optional[Faif source_vision_frames:for source_vision_frame in source_vision_frames:- temp_faces = get_many_faces([source_vision_frame])+ temp_faces = get_static_faces([ source_vision_frame ])temp_faces = sort_faces_by_order(temp_faces, 'large-small')if temp_faces:source_faces.append(get_first(temp_faces))- return get_average_face(source_faces)+ return average_face_identity(source_faces)def process_frame(inputs : FaceSwapperInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')source_vision_frames = inputs.get('source_vision_frames')- target_vision_frame = inputs.get('target_vision_frame')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')++ target_vision_frame = get_middle(target_vision_frames)source_face = extract_source_face(source_vision_frames)- target_faces = select_faces(reference_vision_frame, target_vision_frame)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if source_face and target_faces:+ source_vision_frame = get_first(source_vision_frames)+for target_face in target_faces:target_face = scale_face(target_face, target_vision_frame, temp_vision_frame)- temp_vision_frame = swap_face(source_face, target_face, temp_vision_frame)+ temp_vision_frame = swap_face(source_face, target_face, source_vision_frame, temp_vision_frame)return temp_vision_frame, temp_vision_maskdiff --git a/facefusion/processors/modules/face_swapper/types.py b/facefusion/processors/modules/face_swapper/types.pyindex 4afc3cf5..addda8de 100644--- a/facefusion/processors/modules/face_swapper/types.py+++ b/facefusion/processors/modules/face_swapper/types.py@@ -6,7 +6,7 @@ FaceSwapperInputs = TypedDict('FaceSwapperInputs',{'reference_vision_frame' : VisionFrame,'source_vision_frames' : List[VisionFrame],- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/frame_colorizer/core.py b/facefusion/processors/modules/frame_colorizer/core.pyindex d10ed18b..f5bcc623 100644--- a/facefusion/processors/modules/frame_colorizer/core.py+++ b/facefusion/processors/modules/frame_colorizer/core.py@@ -1,10 +1,12 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleTypefrom typing import Listimport cv2import numpy+import facefusion.choicesimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager@@ -17,8 +19,8 @@ from facefusion.processors.modules.frame_colorizer.types import FrameColorizerInfrom facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import thread_semaphore-from facefusion.types import ApplyStateItem, Args, DownloadScope, ExecutionProvider, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import blend_frame, read_static_image, read_static_video_frame, unpack_resolution+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame+from facefusion.vision import blend_frame, read_static_image, read_static_video_chunk, read_static_video_frame, unpack_resolution@lru_cache()@@ -170,10 +172,11 @@ def clear_inference_pool() -> None:inference_manager.clear_inference_pool(__name__, model_names)-def resolve_execution_providers() -> List[ExecutionProvider]:+def resolve_inference_providers() -> List[InferenceProvider]:if is_macos() and has_execution_provider('coreml'):- return [ 'cpu' ]- return state_manager.get_item('execution_providers')+ return [ facefusion.choices.execution_provider_set.get('cpu') ]++ return []def get_model_options() -> ModelOptions:@@ -196,10 +199,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('frame_colorizer_size', args.get('frame_colorizer_size'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -219,11 +230,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def colorize_frame(temp_vision_frame : VisionFrame) -> VisionFrame:diff --git a/facefusion/processors/modules/frame_colorizer/types.py b/facefusion/processors/modules/frame_colorizer/types.pyindex b0e152f5..8670646d 100644--- a/facefusion/processors/modules/frame_colorizer/types.py+++ b/facefusion/processors/modules/frame_colorizer/types.py@@ -1,10 +1,10 @@-from typing import Literal, TypedDict+from typing import List, Literal, TypedDictfrom facefusion.types import Mask, VisionFrameFrameColorizerInputs = TypedDict('FrameColorizerInputs',{- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/frame_enhancer/core.py b/facefusion/processors/modules/frame_enhancer/core.pyindex fdd980e3..36461e63 100644--- a/facefusion/processors/modules/frame_enhancer/core.py+++ b/facefusion/processors/modules/frame_enhancer/core.py@@ -1,9 +1,12 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleType+from typing import Listimport cv2import numpy+import facefusion.choicesimport facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, inference_manager, logger, state_manager, translator, video_manager@@ -16,8 +19,8 @@ from facefusion.processors.modules.frame_enhancer.types import FrameEnhancerInpufrom facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import conditional_thread_semaphore-from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import blend_frame, create_tile_frames, merge_tile_frames, read_static_image, read_static_video_frame+from facefusion.types import ApplyStateItem, Args, DownloadScope, InferencePool, InferenceProvider, ModelOptions, ModelSet, ProcessMode, VisionFrame+from facefusion.vision import blend_frame, create_tile_frames, merge_tile_frames, read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -156,6 +159,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/real_esrgan_x2_fp16.onnx')}},+ 'precision': 'fp16','size': (256, 16, 8),'scale': 2},@@ -210,6 +214,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/real_esrgan_x4_fp16.onnx')}},+ 'precision': 'fp16','size': (256, 16, 8),'scale': 4},@@ -264,6 +269,7 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:'path': resolve_relative_path('../.assets/models/real_esrgan_x8_fp16.onnx')}},+ 'precision': 'fp16','size': (256, 16, 8),'scale': 8},@@ -541,33 +547,36 @@ def create_static_model_set(download_scope : DownloadScope) -> ModelSet:def get_inference_pool() -> InferencePool:- model_names = [ get_frame_enhancer_model() ]+ model_names = [ state_manager.get_item('frame_enhancer_model') ]model_source_set = get_model_options().get('sources')return inference_manager.get_inference_pool(__name__, model_names, model_source_set)def clear_inference_pool() -> None:- model_names = [ get_frame_enhancer_model() ]+ model_names = [ state_manager.get_item('frame_enhancer_model') ]inference_manager.clear_inference_pool(__name__, model_names)-def get_model_options() -> ModelOptions:- model_name = get_frame_enhancer_model()- return create_static_model_set('full').get(model_name)+def resolve_inference_providers() -> List[InferenceProvider]:+ model_precision = get_model_options().get('precision')++ if is_macos() and has_execution_provider('coreml') and model_precision == 'fp16':+ return\+ [+ (facefusion.choices.execution_provider_set.get('coreml'),+ {+ 'ModelFormat': 'MLProgram',+ 'SpecializationStrategy': 'FastPrediction'+ })+ ]+ return []-def get_frame_enhancer_model() -> str:- frame_enhancer_model = state_manager.get_item('frame_enhancer_model')- if is_macos() and has_execution_provider('coreml'):- if frame_enhancer_model == 'real_esrgan_x2_fp16':- return 'real_esrgan_x2'- if frame_enhancer_model == 'real_esrgan_x4_fp16':- return 'real_esrgan_x4'- if frame_enhancer_model == 'real_esrgan_x8_fp16':- return 'real_esrgan_x8'- return frame_enhancer_model+def get_model_options() -> ModelOptions:+ model_name = state_manager.get_item('frame_enhancer_model')+ return create_static_model_set('full').get(model_name)def register_args(program : ArgumentParser) -> None:@@ -583,10 +592,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('frame_enhancer_blend', args.get('frame_enhancer_blend'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -606,11 +623,15 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def enhance_frame(temp_vision_frame : VisionFrame) -> VisionFrame:diff --git a/facefusion/processors/modules/frame_enhancer/types.py b/facefusion/processors/modules/frame_enhancer/types.pyindex d9786a39..d593b5fb 100644--- a/facefusion/processors/modules/frame_enhancer/types.py+++ b/facefusion/processors/modules/frame_enhancer/types.py@@ -1,10 +1,10 @@-from typing import Literal, TypedDict+from typing import List, Literal, TypedDictfrom facefusion.types import Mask, VisionFrameFrameEnhancerInputs = TypedDict('FrameEnhancerInputs',{- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/processors/modules/lip_syncer/core.py b/facefusion/processors/modules/lip_syncer/core.pyindex b2b73f7b..914288a8 100755--- a/facefusion/processors/modules/lip_syncer/core.py+++ b/facefusion/processors/modules/lip_syncer/core.py@@ -1,5 +1,7 @@from argparse import ArgumentParserfrom functools import lru_cache+from types import ModuleType+from typing import Listimport cv2import numpy@@ -8,9 +10,9 @@ import facefusion.jobs.job_managerimport facefusion.jobs.job_storefrom facefusion import config, content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, inference_manager, logger, state_manager, translator, video_manager, voice_extractorfrom facefusion.audio import read_static_voice-from facefusion.common_helper import create_float_metavar+from facefusion.common_helper import create_float_metavar, get_middlefrom facefusion.download import conditional_download_hashes, conditional_download_sources, resolve_download_url-from facefusion.face_analyser import scale_face+from facefusion.face_creator import scale_facefrom facefusion.face_helper import create_bounding_box, paste_back, warp_face_by_bounding_box, warp_face_by_face_landmark_5from facefusion.face_masker import create_area_mask, create_box_mask, create_occlusion_maskfrom facefusion.face_selector import select_faces@@ -21,7 +23,7 @@ from facefusion.processors.types import ProcessorOutputsfrom facefusion.program_helper import find_argument_groupfrom facefusion.thread_helper import conditional_thread_semaphorefrom facefusion.types import ApplyStateItem, Args, AudioFrame, DownloadScope, Face, InferencePool, ModelOptions, ModelSet, ProcessMode, VisionFrame-from facefusion.vision import read_static_image, read_static_video_frame+from facefusion.vision import read_static_image, read_static_video_chunk, read_static_video_frame@lru_cache()@@ -142,10 +144,18 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:apply_state_item('lip_syncer_weight', args.get('lip_syncer_weight'))+def get_common_modules() -> List[ModuleType]:+ return [ content_analyser, face_classifier, face_detector, face_landmarker, face_masker, face_recognizer, voice_extractor ]++def pre_check() -> bool:model_hash_set = get_model_options().get('hashes')model_source_set = get_model_options().get('sources')+ for common_module in get_common_modules():+ if not common_module.pre_check():+ return False+return conditional_download_hashes(model_hash_set) and conditional_download_sources(model_source_set)@@ -159,18 +169,16 @@ def pre_process(mode : ProcessMode) -> bool:def post_process() -> None:read_static_image.cache_clear()read_static_video_frame.cache_clear()+ read_static_video_chunk.cache_clear()read_static_voice.cache_clear()video_manager.clear_video_pool()+if state_manager.get_item('video_memory_strategy') in [ 'strict', 'moderate' ]:clear_inference_pool()+if state_manager.get_item('video_memory_strategy') == 'strict':- content_analyser.clear_inference_pool()- face_classifier.clear_inference_pool()- face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_masker.clear_inference_pool()- face_recognizer.clear_inference_pool()- voice_extractor.clear_inference_pool()+ for common_module in get_common_modules():+ common_module.clear_inference_pool()def sync_lip(target_face : Face, source_voice_frame : AudioFrame, temp_vision_frame : VisionFrame) -> VisionFrame:@@ -282,11 +290,14 @@ def normalize_crop_frame(crop_vision_frame : VisionFrame) -> VisionFrame:def process_frame(inputs : LipSyncerInputs) -> ProcessorOutputs:reference_vision_frame = inputs.get('reference_vision_frame')+ source_vision_frames = inputs.get('source_vision_frames')source_voice_frame = inputs.get('source_voice_frame')- target_vision_frame = inputs.get('target_vision_frame')+ target_vision_frames = inputs.get('target_vision_frames')temp_vision_frame = inputs.get('temp_vision_frame')temp_vision_mask = inputs.get('temp_vision_mask')- target_faces = select_faces(reference_vision_frame, target_vision_frame)++ target_vision_frame = get_middle(target_vision_frames)+ target_faces = select_faces(reference_vision_frame, source_vision_frames, target_vision_frames)if target_faces:for target_face in target_faces:diff --git a/facefusion/processors/modules/lip_syncer/types.py b/facefusion/processors/modules/lip_syncer/types.pyindex 32861f6a..d13b6a9a 100644--- a/facefusion/processors/modules/lip_syncer/types.py+++ b/facefusion/processors/modules/lip_syncer/types.py@@ -1,4 +1,4 @@-from typing import Any, Literal, TypeAlias, TypedDict+from typing import Any, List, Literal, TypeAlias, TypedDictfrom numpy.typing import NDArray@@ -7,8 +7,9 @@ from facefusion.types import AudioFrame, Mask, VisionFrameLipSyncerInputs = TypedDict('LipSyncerInputs',{'reference_vision_frame' : VisionFrame,+ 'source_vision_frames' : List[VisionFrame],'source_voice_frame' : AudioFrame,- 'target_vision_frame' : VisionFrame,+ 'target_vision_frames' : List[VisionFrame],'temp_vision_frame' : VisionFrame,'temp_vision_mask' : Mask})diff --git a/facefusion/program.py b/facefusion/program.pyindex d029eac9..7d56b006 100755--- a/facefusion/program.py+++ b/facefusion/program.py@@ -133,6 +133,14 @@ def create_face_selector_program() -> ArgumentParser:return program+def create_face_tracker_program() -> ArgumentParser:+ program = ArgumentParser(add_help = False)+ group_face_tracker = program.add_argument_group('face tracker')+ group_face_tracker.add_argument('--face-tracker-score', help = translator.get('help.face_tracker_score'), type = float, default = config.get_float_value('face_tracker', 'face_tracker_score', '0.0'), choices = facefusion.choices.face_tracker_score_range, metavar = create_float_metavar(facefusion.choices.face_tracker_score_range))+ job_store.register_step_keys([ 'face_tracker_score' ])+ return program++def create_face_masker_program() -> ArgumentParser:program = ArgumentParser(add_help = False)group_face_masker = program.add_argument_group('face masker')@@ -166,6 +174,14 @@ def create_frame_extraction_program() -> ArgumentParser:return program+def create_frame_process_program() -> ArgumentParser:+ program = ArgumentParser(add_help = False)+ group_frame_process = program.add_argument_group('frame process')+ group_frame_process.add_argument('--target-frame-amount', help = translator.get('help.target_frame_amount'), type = int, default = config.get_int_value('frame_process', 'target_frame_amount', '5'), choices = facefusion.choices.target_frame_amount_range, metavar = create_int_metavar(facefusion.choices.target_frame_amount_range))+ job_store.register_step_keys([ 'target_frame_amount' ])+ return program++def create_output_creation_program() -> ArgumentParser:program = ArgumentParser(add_help = False)available_encoder_set = get_available_encoder_set()@@ -245,8 +261,7 @@ def create_memory_program() -> ArgumentParser:program = ArgumentParser(add_help = False)group_memory = program.add_argument_group('memory')group_memory.add_argument('--video-memory-strategy', help = translator.get('help.video_memory_strategy'), default = config.get_str_value('memory', 'video_memory_strategy', 'strict'), choices = facefusion.choices.video_memory_strategies)- group_memory.add_argument('--system-memory-limit', help = translator.get('help.system_memory_limit'), type = int, default = config.get_int_value('memory', 'system_memory_limit', '0'), choices = facefusion.choices.system_memory_limit_range, metavar = create_int_metavar(facefusion.choices.system_memory_limit_range))- job_store.register_job_keys([ 'video_memory_strategy', 'system_memory_limit' ])+ job_store.register_job_keys([ 'video_memory_strategy' ])return program@@ -285,7 +300,7 @@ def create_step_index_program() -> ArgumentParser:def collect_step_program() -> ArgumentParser:- return ArgumentParser(parents = [ create_face_detector_program(), create_face_landmarker_program(), create_face_selector_program(), create_face_masker_program(), create_voice_extractor_program(), create_frame_extraction_program(), create_output_creation_program(), create_processors_program() ], add_help = False)+ return ArgumentParser(parents = [ create_face_detector_program(), create_face_landmarker_program(), create_face_selector_program(), create_face_tracker_program(), create_face_masker_program(), create_voice_extractor_program(), create_frame_extraction_program(), create_frame_process_program(), create_output_creation_program(), create_processors_program() ], add_help = False)def collect_job_program() -> ArgumentParser:diff --git a/facefusion/sanitizer.py b/facefusion/sanitizer.pyindex e8c7a920..3f85e532 100644--- a/facefusion/sanitizer.py+++ b/facefusion/sanitizer.py@@ -9,6 +9,7 @@ def sanitize_job_id(job_id : str) -> str:if __job_id__.isalnum():return job_id+return hashlib.sha1(job_id.encode()).hexdigest()diff --git a/facefusion/streamer.py b/facefusion/streamer.pyindex dccb601d..8b91db72 100644--- a/facefusion/streamer.py+++ b/facefusion/streamer.py@@ -2,7 +2,7 @@ import osimport subprocessfrom collections import dequefrom concurrent.futures import ThreadPoolExecutor-from typing import Deque, Iterator+from typing import Deque, Iterator, Listimport cv2import numpy@@ -20,6 +20,7 @@ from facefusion.vision import extract_vision_mask, read_static_imagesdef multi_process_capture(camera_capture : cv2.VideoCapture, camera_fps : Fps) -> Iterator[VisionFrame]:capture_deque : Deque[VisionFrame] = deque()+ source_vision_frames = read_static_images(state_manager.get_item('source_paths'))with tqdm(desc = translator.get('streaming'), unit = 'frame', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:with ThreadPoolExecutor(max_workers = state_manager.get_item('execution_thread_count')) as executor:@@ -31,7 +32,7 @@ def multi_process_capture(camera_capture : cv2.VideoCapture, camera_fps : Fps) -camera_capture.release()if numpy.any(capture_vision_frame):- future = executor.submit(process_stream_frame, capture_vision_frame)+ future = executor.submit(process_stream_frame, source_vision_frames, capture_vision_frame)futures.append(future)for future_done in [ future for future in futures if future.done() ]:@@ -44,8 +45,7 @@ def multi_process_capture(camera_capture : cv2.VideoCapture, camera_fps : Fps) -yield capture_deque.popleft()-def process_stream_frame(target_vision_frame : VisionFrame) -> VisionFrame:- source_vision_frames = read_static_images(state_manager.get_item('source_paths'))+def process_stream_frame(source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame) -> VisionFrame:source_audio_frame = create_empty_audio_frame()source_voice_frame = create_empty_audio_frame()temp_vision_frame = target_vision_frame.copy()@@ -60,7 +60,7 @@ def process_stream_frame(target_vision_frame : VisionFrame) -> VisionFrame:'source_vision_frames': source_vision_frames,'source_audio_frame': source_audio_frame,'source_voice_frame': source_voice_frame,- 'target_vision_frame': target_vision_frame,+ 'target_vision_frames': [ target_vision_frame ],'temp_vision_frame': temp_vision_frame,'temp_vision_mask': temp_vision_mask})diff --git a/facefusion/temp_helper.py b/facefusion/temp_helper.pyindex 9622621b..33d69a94 100644--- a/facefusion/temp_helper.py+++ b/facefusion/temp_helper.py@@ -1,8 +1,8 @@import os-from typing import Listfrom facefusion import state_managerfrom facefusion.filesystem import create_directory, get_file_extension, get_file_name, move_file, remove_directory, resolve_file_pattern+from facefusion.types import FrameSetdef get_temp_file_path(file_path : str) -> str:@@ -16,12 +16,18 @@ def move_temp_file(file_path : str, move_path : str) -> bool:return move_file(temp_file_path, move_path)-def resolve_temp_frame_paths(target_path : str) -> List[str]:- temp_frames_pattern = get_temp_frames_pattern(target_path, '*')- return resolve_file_pattern(temp_frames_pattern)+def resolve_temp_frame_set(target_path : str) -> FrameSet:+ temp_frame_pattern = get_temp_frame_pattern(target_path, '*')+ temp_frame_set = {}+ for temp_frame_path in resolve_file_pattern(temp_frame_pattern):+ frame_number = int(get_file_name(temp_frame_path))+ temp_frame_set[frame_number] = temp_frame_path-def get_temp_frames_pattern(target_path : str, temp_frame_prefix : str) -> str:+ return temp_frame_set+++def get_temp_frame_pattern(target_path : str, temp_frame_prefix : str) -> str:temp_directory_path = get_temp_directory_path(target_path)return os.path.join(temp_directory_path, temp_frame_prefix + '.' + state_manager.get_item('temp_frame_format'))diff --git a/facefusion/types.py b/facefusion/types.pyindex f32b5359..1b1a1a81 100755--- a/facefusion/types.py+++ b/facefusion/types.py@@ -1,5 +1,6 @@from collections import namedtuple-from typing import Any, Callable, Dict, List, Literal, Optional, Tuple, TypeAlias, TypedDict+from threading import Lock+from typing import Any, Callable, Dict, List, Literal, NotRequired, Optional, Tuple, TypeAlias, TypedDictimport cv2import numpy@@ -29,26 +30,34 @@ FaceScoreSet = TypedDict('FaceScoreSet','landmarker' : Score})Embedding : TypeAlias = NDArray[numpy.float64]-Gender = Literal['female', 'male']+Age : TypeAlias = range+Gender = Literal['female', 'male']Race = Literal['white', 'black', 'latino', 'asian', 'indian', 'arabic']++FaceSelectorGender = Literal['auto', 'female', 'male']+FaceSelectorRace = Literal['auto', 'white', 'black', 'latino', 'asian', 'indian', 'arabic']+Face = namedtuple('Face',[+ 'origin','bounding_box','score_set','landmark_set','angle','embedding','embedding_norm',- 'gender','age',+ 'gender','race'])-FaceSet : TypeAlias = Dict[str, List[Face]]-FaceStore = TypedDict('FaceStore',+FaceSet = TypedDict('FaceSet',{- 'static_faces' : FaceSet+ 'lock': Lock,+ 'faces': NotRequired[List[Face]]})+FaceStore : TypeAlias = Dict[str, FaceSet]+FaceTrack : TypeAlias = Dict[int, Face]Language = Literal['en']Locales : TypeAlias = Dict[Language, Dict[str, Any]]@@ -59,12 +68,12 @@ VideoWriterSet : TypeAlias = Dict[str, cv2.VideoWriter]CameraCaptureSet : TypeAlias = Dict[str, cv2.VideoCapture]VideoPoolSet = TypedDict('VideoPoolSet',{- 'capture': VideoCaptureSet,- 'writer': VideoWriterSet+ 'capture' : VideoCaptureSet,+ 'writer' : VideoWriterSet})CameraPoolSet = TypedDict('CameraPoolSet',{- 'capture': CameraCaptureSet+ 'capture' : CameraCaptureSet})ColorMode = Literal['rgb', 'rgba']@@ -138,6 +147,8 @@ AudioTypeSet : TypeAlias = Dict[AudioFormat, str]ImageTypeSet : TypeAlias = Dict[ImageFormat, str]VideoTypeSet : TypeAlias = Dict[VideoFormat, str]+FrameSet : TypeAlias = Dict[int, str]+AudioEncoder = Literal['flac', 'aac', 'libmp3lame', 'libopus', 'libvorbis', 'pcm_s16le', 'pcm_s32le']VideoEncoder = Literal['libx264', 'libx264rgb', 'libx265', 'libvpx-vp9', 'h264_nvenc', 'hevc_nvenc', 'h264_amf', 'hevc_amf', 'h264_qsv', 'hevc_qsv', 'h264_videotoolbox', 'hevc_videotoolbox', 'rawvideo']EncoderSet = TypedDict('EncoderSet',@@ -290,6 +301,7 @@ StateKey = Literal\'reference_face_position','reference_face_distance','reference_frame_number',+ 'face_tracker_score','face_occluder_model','face_parser_model','face_mask_types',@@ -302,6 +314,7 @@ StateKey = Literal\'trim_frame_end','temp_frame_format','keep_temp',+ 'target_frame_amount','output_image_quality','output_image_scale','output_audio_encoder',@@ -320,7 +333,6 @@ StateKey = Literal\'execution_providers','execution_thread_count','video_memory_strategy',- 'system_memory_limit','log_level','halt_on_error','job_id',@@ -346,20 +358,21 @@ State = TypedDict('State','benchmark_cycle_count' : int,'face_detector_model' : FaceDetectorModel,'face_detector_size' : str,- 'face_detector_margin': Margin,+ 'face_detector_margin' : Margin,'face_detector_angles' : List[Angle],'face_detector_score' : Score,'face_landmarker_model' : FaceLandmarkerModel,'face_landmarker_score' : Score,'face_selector_mode' : FaceSelectorMode,'face_selector_order' : FaceSelectorOrder,- 'face_selector_race' : Race,- 'face_selector_gender' : Gender,+ 'face_selector_race' : FaceSelectorRace,+ 'face_selector_gender' : FaceSelectorGender,'face_selector_age_start' : int,'face_selector_age_end' : int,'reference_face_position' : int,'reference_face_distance' : float,'reference_frame_number' : int,+ 'face_tracker_score' : Score,'face_occluder_model' : FaceOccluderModel,'face_parser_model' : FaceParserModel,'face_mask_types' : List[FaceMaskType],@@ -367,11 +380,12 @@ State = TypedDict('State','face_mask_regions' : List[FaceMaskRegion],'face_mask_blur' : float,'face_mask_padding' : Padding,- 'voice_extractor_model': VoiceExtractorModel,+ 'voice_extractor_model' : VoiceExtractorModel,'trim_frame_start' : int,'trim_frame_end' : int,'temp_frame_format' : TempFrameFormat,'keep_temp' : bool,+ 'target_frame_amount' : int,'output_image_quality' : int,'output_image_scale' : Scale,'output_audio_encoder' : AudioEncoder,@@ -390,7 +404,6 @@ State = TypedDict('State','execution_providers' : List[ExecutionProvider],'execution_thread_count' : int,'video_memory_strategy' : VideoMemoryStrategy,- 'system_memory_limit' : int,'log_level' : LogLevel,'halt_on_error' : bool,'job_id' : str,diff --git a/facefusion/uis/assets/overrides.css b/facefusion/uis/assets/overrides.cssindex 0141d598..71d6022e 100644--- a/facefusion/uis/assets/overrides.css+++ b/facefusion/uis/assets/overrides.css@@ -62,10 +62,14 @@width: 1.125rem;}-:root:root:root:root .thumbnail-item+:root:root:root:root .gallery-container .thumbnail-item{border: unset;- box-shadow: unset;+}++:root:root:root:root .gallery-container .grid-container+{+ grid-template-columns: repeat(7, 1fr);}:root:root:root:root .grid-wrap.fixed-heightdiff --git a/facefusion/uis/components/face_selector.py b/facefusion/uis/components/face_selector.pyindex 6a98e7f9..ea744f93 100644--- a/facefusion/uis/components/face_selector.py+++ b/facefusion/uis/components/face_selector.py@@ -7,15 +7,15 @@ from gradio_rangeslider import RangeSliderimport facefusion.choicesfrom facefusion import state_manager, translatorfrom facefusion.common_helper import calculate_float_step, calculate_int_step-from facefusion.face_analyser import get_many_faces+from facefusion.face_creator import get_many_facesfrom facefusion.face_selector import sort_and_filter_faces-from facefusion.face_store import clear_static_faces-from facefusion.filesystem import is_image, is_video-from facefusion.types import FaceSelectorMode, FaceSelectorOrder, Gender, Race, VisionFrame+from facefusion.face_store import clear_faces+from facefusion.filesystem import filter_image_paths, is_image, is_video+from facefusion.types import FaceSelectorGender, FaceSelectorMode, FaceSelectorOrder, FaceSelectorRace, VisionFramefrom facefusion.uis.core import get_ui_component, get_ui_components, register_ui_componentfrom facefusion.uis.types import ComponentOptionsfrom facefusion.uis.ui_helper import convert_str_none-from facefusion.vision import fit_cover_frame, read_static_image, read_video_frame+from facefusion.vision import fit_cover_frame, read_static_image, read_static_images, read_video_frameFACE_SELECTOR_MODE_DROPDOWN : Optional[gradio.Dropdown] = NoneFACE_SELECTOR_ORDER_DROPDOWN : Optional[gradio.Dropdown] = None@@ -39,17 +39,18 @@ def render() -> None:{'label': translator.get('uis.reference_face_gallery'),'object_fit': 'cover',- 'columns': 7,'allow_preview': False,'elem_classes': 'box-face-selector','visible': 'reference' in state_manager.get_item('face_selector_mode')}+ source_vision_frames = read_static_images(filter_image_paths(state_manager.get_item('source_paths')))+if is_image(state_manager.get_item('target_path')):target_vision_frame = read_static_image(state_manager.get_item('target_path'))- reference_face_gallery_options['value'] = extract_gallery_frames(target_vision_frame)+ reference_face_gallery_options['value'] = extract_gallery_frames(source_vision_frames, target_vision_frame)if is_video(state_manager.get_item('target_path')):target_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))- reference_face_gallery_options['value'] = extract_gallery_frames(target_vision_frame)+ reference_face_gallery_options['value'] = extract_gallery_frames(source_vision_frames, target_vision_frame)FACE_SELECTOR_MODE_DROPDOWN = gradio.Dropdown(label = translator.get('uis.face_selector_mode_dropdown'),choices = facefusion.choices.face_selector_modes,@@ -156,12 +157,12 @@ def update_face_selector_order(face_analyser_order : FaceSelectorOrder) -> gradireturn update_reference_position_gallery()-def update_face_selector_gender(face_selector_gender : Gender) -> gradio.Gallery:+def update_face_selector_gender(face_selector_gender : FaceSelectorGender) -> gradio.Gallery:state_manager.set_item('face_selector_gender', convert_str_none(face_selector_gender))return update_reference_position_gallery()-def update_face_selector_race(face_selector_race : Race) -> gradio.Gallery:+def update_face_selector_race(face_selector_race : FaceSelectorRace) -> gradio.Gallery:state_manager.set_item('face_selector_race', convert_str_none(face_selector_race))return update_reference_position_gallery()@@ -194,30 +195,33 @@ def clear_reference_frame_number() -> None:def clear_and_update_reference_position_gallery() -> gradio.Gallery:- clear_static_faces()+ clear_faces()return update_reference_position_gallery()def update_reference_position_gallery(frame_number : int = 0) -> gradio.Gallery:gallery_vision_frames = []+ source_vision_frames = read_static_images(filter_image_paths(state_manager.get_item('source_paths')))+if is_image(state_manager.get_item('target_path')):target_vision_frame = read_static_image(state_manager.get_item('target_path'))- gallery_vision_frames = extract_gallery_frames(target_vision_frame)+ gallery_vision_frames = extract_gallery_frames(source_vision_frames, target_vision_frame)if is_video(state_manager.get_item('target_path')):target_vision_frame = read_video_frame(state_manager.get_item('target_path'), frame_number)- gallery_vision_frames = extract_gallery_frames(target_vision_frame)+ gallery_vision_frames = extract_gallery_frames(source_vision_frames, target_vision_frame)if gallery_vision_frames:return gradio.Gallery(value = gallery_vision_frames)return gradio.Gallery(value = None)-def extract_gallery_frames(target_vision_frame : VisionFrame) -> List[VisionFrame]:+def extract_gallery_frames(source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame) -> List[VisionFrame]:gallery_vision_frames = []- faces = get_many_faces([ target_vision_frame ])- faces = sort_and_filter_faces(faces)+ source_faces = get_many_faces(source_vision_frames)+ target_faces = get_many_faces([ target_vision_frame ])+ target_faces = sort_and_filter_faces(source_faces, target_faces)- for face in faces:- start_x, start_y, end_x, end_y = map(int, face.bounding_box)+ for target_face in target_faces:+ start_x, start_y, end_x, end_y = map(int, target_face.bounding_box)padding_x = int((end_x - start_x) * 0.25)padding_y = int((end_y - start_y) * 0.25)start_x = max(0, start_x - padding_x)@@ -228,4 +232,5 @@ def extract_gallery_frames(target_vision_frame : VisionFrame) -> List[VisionFramcrop_vision_frame = fit_cover_frame(crop_vision_frame, (128, 128))crop_vision_frame = cv2.cvtColor(crop_vision_frame, cv2.COLOR_BGR2RGB)gallery_vision_frames.append(crop_vision_frame)+return gallery_vision_framesdiff --git a/facefusion/uis/components/face_tracker.py b/facefusion/uis/components/face_tracker.pynew file mode 100644index 00000000..2631a26c--- /dev/null+++ b/facefusion/uis/components/face_tracker.py@@ -0,0 +1,32 @@+from typing import Optional++import gradio++import facefusion.choices+from facefusion import state_manager, translator+from facefusion.common_helper import calculate_float_step+from facefusion.types import Score+from facefusion.uis.core import register_ui_component++FACE_TRACKER_SCORE_SLIDER : Optional[gradio.Slider] = None+++def render() -> None:+ global FACE_TRACKER_SCORE_SLIDER++ FACE_TRACKER_SCORE_SLIDER = gradio.Slider(+ label = translator.get('uis.face_tracker_score_slider'),+ value = state_manager.get_item('face_tracker_score'),+ step = calculate_float_step(facefusion.choices.face_tracker_score_range),+ minimum = facefusion.choices.face_tracker_score_range[0],+ maximum = facefusion.choices.face_tracker_score_range[-1]+ )+ register_ui_component('face_tracker_score_slider', FACE_TRACKER_SCORE_SLIDER)+++def listen() -> None:+ FACE_TRACKER_SCORE_SLIDER.release(update_face_tracker_score, inputs = FACE_TRACKER_SCORE_SLIDER)+++def update_face_tracker_score(face_tracker_score : Score) -> None:+ state_manager.set_item('face_tracker_score', face_tracker_score)diff --git a/facefusion/uis/components/memory.py b/facefusion/uis/components/memory.pyindex cbb7b425..07f6b477 100644--- a/facefusion/uis/components/memory.py+++ b/facefusion/uis/components/memory.py@@ -4,39 +4,24 @@ import gradioimport facefusion.choicesfrom facefusion import state_manager, translator-from facefusion.common_helper import calculate_int_stepfrom facefusion.types import VideoMemoryStrategyVIDEO_MEMORY_STRATEGY_DROPDOWN : Optional[gradio.Dropdown] = None-SYSTEM_MEMORY_LIMIT_SLIDER : Optional[gradio.Slider] = Nonedef render() -> None:global VIDEO_MEMORY_STRATEGY_DROPDOWN- global SYSTEM_MEMORY_LIMIT_SLIDERVIDEO_MEMORY_STRATEGY_DROPDOWN = gradio.Dropdown(label = translator.get('uis.video_memory_strategy_dropdown'),choices = facefusion.choices.video_memory_strategies,value = state_manager.get_item('video_memory_strategy'))- SYSTEM_MEMORY_LIMIT_SLIDER = gradio.Slider(- label = translator.get('uis.system_memory_limit_slider'),- step = calculate_int_step(facefusion.choices.system_memory_limit_range),- minimum = facefusion.choices.system_memory_limit_range[0],- maximum = facefusion.choices.system_memory_limit_range[-1],- value = state_manager.get_item('system_memory_limit')- )def listen() -> None:VIDEO_MEMORY_STRATEGY_DROPDOWN.change(update_video_memory_strategy, inputs = VIDEO_MEMORY_STRATEGY_DROPDOWN)- SYSTEM_MEMORY_LIMIT_SLIDER.release(update_system_memory_limit, inputs = SYSTEM_MEMORY_LIMIT_SLIDER)def update_video_memory_strategy(video_memory_strategy : VideoMemoryStrategy) -> None:state_manager.set_item('video_memory_strategy', video_memory_strategy)---def update_system_memory_limit(system_memory_limit : float) -> None:- state_manager.set_item('system_memory_limit', int(system_memory_limit))diff --git a/facefusion/uis/components/preview.py b/facefusion/uis/components/preview.pyindex 69da95fa..5c4de8be 100755--- a/facefusion/uis/components/preview.py+++ b/facefusion/uis/components/preview.py@@ -7,18 +7,18 @@ import numpyfrom facefusion import logger, process_manager, state_manager, translatorfrom facefusion.audio import create_empty_audio_frame, get_voice_frame-from facefusion.common_helper import get_first+from facefusion.common_helper import get_first, get_middlefrom facefusion.content_analyser import analyse_frame-from facefusion.face_analyser import get_one_face+from facefusion.face_creator import get_one_facefrom facefusion.face_selector import select_faces-from facefusion.face_store import clear_static_faces+from facefusion.face_store import clear_facesfrom facefusion.filesystem import filter_audio_paths, is_image, is_videofrom facefusion.processors.core import get_processors_modulesfrom facefusion.types import AudioFrame, Face, Mask, VisionFramefrom facefusion.uis import choices as uis_choicesfrom facefusion.uis.core import get_ui_component, get_ui_components, register_ui_componentfrom facefusion.uis.types import ComponentOptions, PreviewMode-from facefusion.vision import detect_frame_orientation, extract_vision_mask, fit_cover_frame, merge_vision_mask, obscure_frame, read_static_image, read_static_images, read_video_frame, restrict_frame, unpack_resolution+from facefusion.vision import detect_frame_orientation, extract_vision_mask, fit_cover_frame, merge_vision_mask, obscure_frame, read_static_image, read_static_images, read_video_frame, restrict_frame, select_video_frames, unpack_resolutionPREVIEW_IMAGE : Optional[gradio.Image] = None@@ -36,7 +36,7 @@ def render() -> None:source_audio_frame = create_empty_audio_frame()source_voice_frame = create_empty_audio_frame()- if source_audio_path and state_manager.get_item('output_video_fps') and state_manager.get_item('reference_frame_number'):+ if source_audio_path and state_manager.get_item('output_video_fps'):temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), state_manager.get_item('reference_frame_number'))if numpy.any(temp_voice_frame):source_voice_frame = temp_voice_frame@@ -44,14 +44,14 @@ def render() -> None:if is_image(state_manager.get_item('target_path')):target_vision_frame = read_static_image(state_manager.get_item('target_path'))reference_vision_frame = read_static_image(state_manager.get_item('target_path'))- preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frame, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])+ preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, [ target_vision_frame ], uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])preview_image_options['value'] = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGR2RGB)preview_image_options['elem_classes'] = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ]if is_video(state_manager.get_item('target_path')):- temp_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))reference_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))- preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, temp_vision_frame, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])+ target_vision_frames = select_video_frames(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'), state_manager.get_item('target_frame_amount'))+ preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frames, uis_choices.preview_modes[0], uis_choices.preview_resolutions[-1])preview_image_options['value'] = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGR2RGB)preview_image_options['elem_classes'] = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ]preview_image_options['visible'] = True@@ -134,6 +134,7 @@ def listen() -> None:'lip_syncer_weight_slider','reference_face_distance_slider','face_selector_age_range_slider',+ 'face_tracker_score_slider','face_mask_blur_slider','face_mask_padding_top_slider','face_mask_padding_bottom_slider',@@ -189,43 +190,42 @@ def update_preview_image(preview_mode : PreviewMode, preview_resolution : str, fsource_audio_frame = create_empty_audio_frame()source_voice_frame = create_empty_audio_frame()- if source_audio_path and state_manager.get_item('output_video_fps') and state_manager.get_item('reference_frame_number'):- reference_audio_frame_number = state_manager.get_item('reference_frame_number')+ if source_audio_path and state_manager.get_item('output_video_fps'):+ audio_frame_number = frame_numberif state_manager.get_item('trim_frame_start'):- reference_audio_frame_number -= state_manager.get_item('trim_frame_start')- temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), reference_audio_frame_number)+ audio_frame_number -= state_manager.get_item('trim_frame_start')+ temp_voice_frame = get_voice_frame(source_audio_path, state_manager.get_item('output_video_fps'), audio_frame_number)if numpy.any(temp_voice_frame):source_voice_frame = temp_voice_frameif is_image(state_manager.get_item('target_path')):reference_vision_frame = read_static_image(state_manager.get_item('target_path'))target_vision_frame = read_static_image(state_manager.get_item('target_path'), 'rgba')- target_vision_mask = extract_vision_mask(target_vision_frame)- target_vision_frame = merge_vision_mask(target_vision_frame, target_vision_mask)- preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frame, preview_mode, preview_resolution)+ preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, [ target_vision_frame ], preview_mode, preview_resolution)preview_vision_frame = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGRA2RGBA)return gradio.Image(value = preview_vision_frame, elem_classes = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ])if is_video(state_manager.get_item('target_path')):reference_vision_frame = read_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))- temp_vision_frame = read_video_frame(state_manager.get_item('target_path'), frame_number)- temp_vision_mask = extract_vision_mask(temp_vision_frame)- temp_vision_frame = merge_vision_mask(temp_vision_frame, temp_vision_mask)- preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, temp_vision_frame, preview_mode, preview_resolution)+ target_vision_frames = select_video_frames(state_manager.get_item('target_path'), frame_number, state_manager.get_item('target_frame_amount'))+ preview_vision_frame = process_preview_frame(reference_vision_frame, source_vision_frames, source_audio_frame, source_voice_frame, target_vision_frames, preview_mode, preview_resolution)preview_vision_frame = cv2.cvtColor(preview_vision_frame, cv2.COLOR_BGRA2RGBA)return gradio.Image(value = preview_vision_frame, elem_classes = [ 'image-preview', 'is-' + detect_frame_orientation(preview_vision_frame) ])return gradio.Image(value = None, elem_classes = None)def clear_and_update_preview_image(preview_mode : PreviewMode, preview_resolution : str, frame_number : int = 0) -> gradio.Image:- clear_static_faces()+ clear_faces()return update_preview_image(preview_mode, preview_resolution, frame_number)-def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], source_audio_frame : AudioFrame, source_voice_frame : AudioFrame, target_vision_frame : VisionFrame, preview_mode : PreviewMode, preview_resolution : str) -> VisionFrame:+def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], source_audio_frame : AudioFrame, source_voice_frame : AudioFrame, target_vision_frames : List[VisionFrame], preview_mode : PreviewMode, preview_resolution : str) -> VisionFrame:+ target_vision_frame = get_middle(target_vision_frames)target_vision_frame = restrict_frame(target_vision_frame, unpack_resolution(preview_resolution))+ temp_vision_mask = extract_vision_mask(target_vision_frame)+ target_vision_frame = merge_vision_mask(target_vision_frame, temp_vision_mask)+ target_vision_frames = [ restrict_frame(vision_frame, unpack_resolution(preview_resolution))[:, :, :3] for vision_frame in target_vision_frames ]temp_vision_frame = target_vision_frame.copy()- temp_vision_mask = extract_vision_mask(temp_vision_frame)if analyse_frame(target_vision_frame[:, :, :3]):if preview_mode == 'frame-by-frame':@@ -233,7 +233,7 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frreturn numpy.hstack((temp_vision_frame, temp_vision_frame))if preview_mode == 'face-by-face':- target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, target_vision_frame[:, :, :3], temp_vision_frame[:, :, :3])+ target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, source_vision_frames, target_vision_frame[:, :, :3], temp_vision_frame[:, :, :3])target_crop_vision_frame = obscure_frame(target_crop_vision_frame)output_crop_vision_frame = obscure_frame(output_crop_vision_frame)return numpy.hstack((target_crop_vision_frame, output_crop_vision_frame))@@ -251,7 +251,7 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_fr'source_audio_frame': source_audio_frame,'source_voice_frame': source_voice_frame,'source_vision_frames': source_vision_frames,- 'target_vision_frame': target_vision_frame[:, :, :3],+ 'target_vision_frames': target_vision_frames,'temp_vision_frame': temp_vision_frame[:, :, :3],'temp_vision_mask': temp_vision_mask})@@ -263,14 +263,14 @@ def process_preview_frame(reference_vision_frame : VisionFrame, source_vision_frreturn numpy.hstack((target_vision_frame, temp_vision_frame))if preview_mode == 'face-by-face':- target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, target_vision_frame, temp_vision_frame)+ target_crop_vision_frame, output_crop_vision_frame = create_face_by_face(reference_vision_frame, source_vision_frames, target_vision_frame, temp_vision_frame)return numpy.hstack((target_crop_vision_frame, output_crop_vision_frame))return temp_vision_frame-def create_face_by_face(reference_vision_frame : VisionFrame, target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, VisionFrame]:- target_faces = select_faces(reference_vision_frame[:, :, :3], target_vision_frame[:, :, :3])+def create_face_by_face(reference_vision_frame : VisionFrame, source_vision_frames : List[VisionFrame], target_vision_frame : VisionFrame, temp_vision_frame : VisionFrame) -> Tuple[VisionFrame, VisionFrame]:+ target_faces = select_faces(reference_vision_frame[:, :, :3], source_vision_frames, [ target_vision_frame[:, :, :3] ])target_face = get_one_face(target_faces)if target_face:diff --git a/facefusion/uis/components/target.py b/facefusion/uis/components/target.pyindex 405e294d..e69e3c56 100644--- a/facefusion/uis/components/target.py+++ b/facefusion/uis/components/target.py@@ -3,7 +3,7 @@ from typing import Optional, Tupleimport gradiofrom facefusion import state_manager, translator-from facefusion.face_store import clear_static_faces+from facefusion.face_store import clear_facesfrom facefusion.filesystem import is_image, is_videofrom facefusion.uis.core import register_ui_componentfrom facefusion.uis.types import ComponentOptions, File@@ -51,7 +51,7 @@ def listen() -> None:def update(file : File) -> Tuple[gradio.Image, gradio.Video]:- clear_static_faces()+ clear_faces()if file and is_image(file.name):state_manager.set_item('target_path', file.name)diff --git a/facefusion/uis/components/trim_frame.py b/facefusion/uis/components/trim_frame.pyindex 8c5eb7db..9ad37a6e 100644--- a/facefusion/uis/components/trim_frame.py+++ b/facefusion/uis/components/trim_frame.py@@ -3,7 +3,7 @@ from typing import Optional, Tuplefrom gradio_rangeslider import RangeSliderfrom facefusion import state_manager, translator-from facefusion.face_store import clear_static_faces+from facefusion.face_store import clear_facesfrom facefusion.filesystem import is_videofrom facefusion.uis.core import get_ui_componentsfrom facefusion.uis.types import ComponentOptions@@ -53,7 +53,7 @@ def remote_update() -> RangeSlider:def update_trim_frame(trim_frame : Tuple[float, float]) -> None:- clear_static_faces()+ clear_faces()trim_frame_start, trim_frame_end = trim_framevideo_frame_total = count_video_frame_total(state_manager.get_item('target_path'))trim_frame_start = int(trim_frame_start) if trim_frame_start > 0 else Nonediff --git a/facefusion/uis/components/voice_extractor.py b/facefusion/uis/components/voice_extractor.pyindex a0845b45..7ea3893c 100644--- a/facefusion/uis/components/voice_extractor.py+++ b/facefusion/uis/components/voice_extractor.py@@ -1,4 +1,4 @@-from typing import Optional+from typing import List, Optionalimport gradio@@ -6,7 +6,7 @@ import facefusion.choicesfrom facefusion import state_manager, translator, voice_extractorfrom facefusion.filesystem import is_videofrom facefusion.types import VoiceExtractorModel-from facefusion.uis.core import get_ui_components, register_ui_component+from facefusion.uis.core import get_ui_component, get_ui_components, register_ui_componentVOICE_EXTRACTOR_MODEL_DROPDOWN : Optional[gradio.Dropdown] = None@@ -14,11 +14,12 @@ VOICE_EXTRACTOR_MODEL_DROPDOWN : Optional[gradio.Dropdown] = Nonedef render() -> None:global VOICE_EXTRACTOR_MODEL_DROPDOWN+ has_lip_syncer = 'lip_syncer' in state_manager.get_item('processors')VOICE_EXTRACTOR_MODEL_DROPDOWN = gradio.Dropdown(label = translator.get('uis.voice_extractor_model_dropdown'),choices = facefusion.choices.voice_extractor_models,value = state_manager.get_item('voice_extractor_model'),- visible = is_video(state_manager.get_item('target_path'))+ visible = is_video(state_manager.get_item('target_path')) and has_lip_syncer)register_ui_component('voice_extractor_model_dropdown', VOICE_EXTRACTOR_MODEL_DROPDOWN)@@ -26,17 +27,22 @@ def render() -> None:def listen() -> None:VOICE_EXTRACTOR_MODEL_DROPDOWN.change(update_voice_extractor_model, inputs = VOICE_EXTRACTOR_MODEL_DROPDOWN, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)- for ui_component in get_ui_components(- [- 'target_image',- 'target_video'- ]):- for method in [ 'change', 'clear' ]:- getattr(ui_component, method)(remote_update, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)+ processors_checkbox_group = get_ui_component('processors_checkbox_group')+ if processors_checkbox_group:+ processors_checkbox_group.change(remote_update, inputs = processors_checkbox_group, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)+ for ui_component in get_ui_components(+ [+ 'target_image',+ 'target_video'+ ]):+ for method in [ 'change', 'clear' ]:+ getattr(ui_component, method)(remote_update, inputs = processors_checkbox_group, outputs = VOICE_EXTRACTOR_MODEL_DROPDOWN)-def remote_update() -> gradio.Dropdown:- if is_video(state_manager.get_item('target_path')):++def remote_update(processors : List[str]) -> gradio.Dropdown:+ has_lip_syncer = 'lip_syncer' in processors+ if is_video(state_manager.get_item('target_path')) and has_lip_syncer:return gradio.Dropdown(visible = True)return gradio.Dropdown(visible = False)diff --git a/facefusion/uis/layouts/default.py b/facefusion/uis/layouts/default.pyindex ef1a0727..cf0aeace 100755--- a/facefusion/uis/layouts/default.py+++ b/facefusion/uis/layouts/default.py@@ -1,7 +1,7 @@import gradiofrom facefusion import state_manager-from facefusion.uis.components import about, age_modifier_options, background_remover_options, common_options, deep_swapper_options, download, execution, execution_thread_count, expression_restorer_options, face_debugger_options, face_detector, face_editor_options, face_enhancer_options, face_landmarker, face_masker, face_selector, face_swapper_options, frame_colorizer_options, frame_enhancer_options, instant_runner, job_manager, job_runner, lip_syncer_options, memory, output, output_options, preview, preview_options, processors, source, target, temp_frame, terminal, trim_frame, ui_workflow, voice_extractor+from facefusion.uis.components import about, age_modifier_options, background_remover_options, common_options, deep_swapper_options, download, execution, execution_thread_count, expression_restorer_options, face_debugger_options, face_detector, face_editor_options, face_enhancer_options, face_landmarker, face_masker, face_selector, face_swapper_options, face_tracker, frame_colorizer_options, frame_enhancer_options, instant_runner, job_manager, job_runner, lip_syncer_options, memory, output, output_options, preview, preview_options, processors, source, target, temp_frame, terminal, trim_frame, ui_workflow, voice_extractordef pre_check() -> bool:@@ -73,6 +73,8 @@ def render() -> gradio.Blocks:trim_frame.render()with gradio.Blocks():face_selector.render()+ with gradio.Blocks():+ face_tracker.render()with gradio.Blocks():face_masker.render()with gradio.Blocks():@@ -114,6 +116,7 @@ def listen() -> None:preview_options.listen()trim_frame.listen()face_selector.listen()+ face_tracker.listen()face_masker.listen()face_detector.listen()face_landmarker.listen()diff --git a/facefusion/uis/types.py b/facefusion/uis/types.pyindex 0e534056..8d9e3bc4 100644--- a/facefusion/uis/types.py+++ b/facefusion/uis/types.py@@ -58,6 +58,7 @@ ComponentName = Literal\'face_selector_mode_dropdown','face_selector_order_dropdown','face_selector_race_dropdown',+ 'face_tracker_score_slider','face_swapper_model_dropdown','face_swapper_pixel_boost_dropdown','face_swapper_weight_slider',diff --git a/facefusion/vision.py b/facefusion/vision.pyindex 561e374c..50b2a8d9 100644--- a/facefusion/vision.py+++ b/facefusion/vision.py@@ -1,6 +1,6 @@import mathfrom functools import lru_cache-from typing import List, Optional, Tuple+from typing import Dict, List, Optional, Tupleimport cv2import numpy@@ -8,7 +8,7 @@ from cv2.typing import Sizefrom facefusion.common_helper import is_windowsfrom facefusion.filesystem import get_file_extension, is_image, is_video-from facefusion.thread_helper import thread_semaphore+from facefusion.thread_helper import thread_lock, thread_semaphorefrom facefusion.types import ColorMode, Duration, Fps, Mask, Orientation, Resolution, Scale, VisionFramefrom facefusion.video_manager import get_video_capture@@ -81,9 +81,10 @@ def read_video_frame(video_path : str, frame_number : int = 0) -> Optional[Visioif video_capture and video_capture.isOpened():frame_total = video_capture.get(cv2.CAP_PROP_FRAME_COUNT)+ frame_position = min(frame_total, frame_number)with thread_semaphore():- video_capture.set(cv2.CAP_PROP_POS_FRAMES, min(frame_total, frame_number - 1))+ video_capture.set(cv2.CAP_PROP_POS_FRAMES, frame_position)has_vision_frame, vision_frame = video_capture.read()if has_vision_frame:@@ -92,6 +93,51 @@ def read_video_frame(video_path : str, frame_number : int = 0) -> Optional[Visioreturn None+@lru_cache(maxsize = 2)+def read_static_video_chunk(video_path : str, chunk_number : int, chunk_size : int) -> Dict[int, VisionFrame]:+ return read_video_chunk(video_path, chunk_number, chunk_size)+++def read_video_chunk(video_path : str, chunk_number : int, chunk_size : int) -> Dict[int, VisionFrame]:+ video_frame_chunk = {}++ if is_video(video_path) and chunk_number > -1:+ video_capture = get_video_capture(video_path)++ if video_capture and video_capture.isOpened():+ frame_total = int(video_capture.get(cv2.CAP_PROP_FRAME_COUNT))+ frame_position = chunk_number * chunk_size++ with thread_semaphore():+ video_capture.set(cv2.CAP_PROP_POS_FRAMES, frame_position)++ for frame_number in range(frame_position, min(frame_position + chunk_size, frame_total)):+ has_vision_frame, vision_frame = video_capture.read()++ if has_vision_frame:+ video_frame_chunk[frame_number] = vision_frame++ return video_frame_chunk+++def select_video_frames(video_path : str, frame_number : int = 0, frame_offset : int = 5) -> List[VisionFrame]:+ vision_frames = []+ chunk_size = (frame_offset * 2 + 1) * 4++ if is_video(video_path):+ with thread_lock():+ for current_number in range(frame_number - frame_offset, frame_number + frame_offset + 1):+ video_frame_chunk = read_static_video_chunk(video_path, current_number // chunk_size, chunk_size)+ vision_frame = create_empty_vision_frame()++ if current_number in video_frame_chunk:+ vision_frame = video_frame_chunk.get(current_number)++ vision_frames.append(vision_frame)++ return vision_frames++def count_video_frame_total(video_path : str) -> int:if is_video(video_path):video_capture = get_video_capture(video_path)@@ -307,6 +353,10 @@ def blend_vision_frames(source_vision_frame : VisionFrame, target_vision_frame :return blend_vision_frame+def create_empty_vision_frame() -> VisionFrame:+ return numpy.zeros((1, 1, 3)).astype(numpy.uint8)++def create_tile_frames(vision_frame : VisionFrame, size : Size) -> Tuple[List[VisionFrame], int, int]:tile_width = size[0] - 2 * size[2]pad_size_top = size[1] + size[2]diff --git a/facefusion/workflows/image_to_image.py b/facefusion/workflows/image_to_image.pyindex ad06f8b4..766a9dd9 100644--- a/facefusion/workflows/image_to_image.py+++ b/facefusion/workflows/image_to_image.py@@ -38,10 +38,12 @@ def setup() -> ErrorCode:if analyse_image(state_manager.get_item('target_path')):return 3- logger.debug(translator.get('clearing_temp'), __name__)- clear_temp_directory(state_manager.get_item('target_path'))- logger.debug(translator.get('creating_temp'), __name__)- create_temp_directory(state_manager.get_item('target_path'))+ if clear_temp_directory(state_manager.get_item('target_path')):+ logger.debug(translator.get('clearing_temp'), __name__)++ if create_temp_directory(state_manager.get_item('target_path')):+ logger.debug(translator.get('creating_temp'), __name__)+return 0@@ -65,8 +67,7 @@ def process_image() -> ErrorCode:source_vision_frames = read_static_images(state_manager.get_item('source_paths'))source_audio_frame = create_empty_audio_frame()source_voice_frame = create_empty_audio_frame()- target_vision_frame = read_static_image(temp_image_path, 'rgba')- temp_vision_frame = target_vision_frame.copy()+ temp_vision_frame = read_static_image(temp_image_path, 'rgba')temp_vision_mask = extract_vision_mask(temp_vision_frame)for processor_module in get_processors_modules(state_manager.get_item('processors')):@@ -78,7 +79,7 @@ def process_image() -> ErrorCode:'source_vision_frames': source_vision_frames,'source_audio_frame': source_audio_frame,'source_voice_frame': source_voice_frame,- 'target_vision_frame': target_vision_frame[:, :, :3],+ 'target_vision_frames': [ temp_vision_frame[:, :, :3] ],'temp_vision_frame': temp_vision_frame[:, :, :3],'temp_vision_mask': temp_vision_mask})diff --git a/facefusion/workflows/image_to_video.py b/facefusion/workflows/image_to_video.pyindex 45220f32..19e6538e 100644--- a/facefusion/workflows/image_to_video.py+++ b/facefusion/workflows/image_to_video.py@@ -11,10 +11,10 @@ from facefusion.common_helper import get_firstfrom facefusion.content_analyser import analyse_videofrom facefusion.filesystem import filter_audio_paths, is_videofrom facefusion.processors.core import get_processors_modules-from facefusion.temp_helper import clear_temp_directory, create_temp_directory, move_temp_file, resolve_temp_frame_paths+from facefusion.temp_helper import clear_temp_directory, create_temp_directory, move_temp_file, resolve_temp_frame_setfrom facefusion.time_helper import calculate_end_timefrom facefusion.types import ErrorCode-from facefusion.vision import conditional_merge_vision_mask, detect_video_resolution, extract_vision_mask, pack_resolution, read_static_image, read_static_images, read_static_video_frame, restrict_trim_frame, restrict_video_fps, restrict_video_resolution, scale_resolution, write_image+from facefusion.vision import conditional_merge_vision_mask, detect_video_resolution, extract_vision_mask, pack_resolution, read_static_image, read_static_images, read_static_video_frame, restrict_trim_frame, restrict_video_fps, restrict_video_resolution, scale_resolution, select_video_frames, write_imagefrom facefusion.workflows.core import is_process_stopping@@ -47,10 +47,12 @@ def setup() -> ErrorCode:if analyse_video(state_manager.get_item('target_path'), trim_frame_start, trim_frame_end):return 3- logger.debug(translator.get('clearing_temp'), __name__)- clear_temp_directory(state_manager.get_item('target_path'))- logger.debug(translator.get('creating_temp'), __name__)- create_temp_directory(state_manager.get_item('target_path'))+ if clear_temp_directory(state_manager.get_item('target_path')):+ logger.debug(translator.get('clearing_temp'), __name__)++ if create_temp_directory(state_manager.get_item('target_path')):+ logger.debug(translator.get('creating_temp'), __name__)+return 0@@ -72,16 +74,16 @@ def extract_frames() -> ErrorCode:def process_video() -> ErrorCode:- temp_frame_paths = resolve_temp_frame_paths(state_manager.get_item('target_path'))+ temp_frame_set = resolve_temp_frame_set(state_manager.get_item('target_path'))- if temp_frame_paths:- with tqdm(total = len(temp_frame_paths), desc = translator.get('processing'), unit = 'frame', ascii = ' =', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:+ if temp_frame_set:+ with tqdm(total = len(temp_frame_set), desc = translator.get('processing'), unit = 'frame', ascii = ' =', disable = state_manager.get_item('log_level') in [ 'warn', 'error' ]) as progress:progress.set_postfix(execution_providers = state_manager.get_item('execution_providers'))with ThreadPoolExecutor(max_workers = state_manager.get_item('execution_thread_count')) as executor:futures = []- for frame_number, temp_frame_path in enumerate(temp_frame_paths):+ for frame_number, temp_frame_path in temp_frame_set.items():future = executor.submit(process_temp_frame, temp_frame_path, frame_number)futures.append(future)@@ -153,16 +155,17 @@ def restore_audio() -> ErrorCode:def process_temp_frame(temp_frame_path : str, frame_number : int) -> bool:+ trim_frame_start, _ = restrict_trim_frame(state_manager.get_item('target_path'), state_manager.get_item('trim_frame_start'), state_manager.get_item('trim_frame_end'))reference_vision_frame = read_static_video_frame(state_manager.get_item('target_path'), state_manager.get_item('reference_frame_number'))source_vision_frames = read_static_images(state_manager.get_item('source_paths'))source_audio_path = get_first(filter_audio_paths(state_manager.get_item('source_paths')))+ target_vision_frames = select_video_frames(state_manager.get_item('target_path'), frame_number, state_manager.get_item('target_frame_amount'))temp_video_fps = restrict_video_fps(state_manager.get_item('target_path'), state_manager.get_item('output_video_fps'))- target_vision_frame = read_static_image(temp_frame_path, 'rgba')- temp_vision_frame = target_vision_frame.copy()+ temp_vision_frame = read_static_image(temp_frame_path, 'rgba')temp_vision_mask = extract_vision_mask(temp_vision_frame)- source_audio_frame = get_audio_frame(source_audio_path, temp_video_fps, frame_number)- source_voice_frame = get_voice_frame(source_audio_path, temp_video_fps, frame_number)+ source_audio_frame = get_audio_frame(source_audio_path, temp_video_fps, frame_number - trim_frame_start)+ source_voice_frame = get_voice_frame(source_audio_path, temp_video_fps, frame_number - trim_frame_start)if not numpy.any(source_audio_frame):source_audio_frame = create_empty_audio_frame()@@ -176,7 +179,7 @@ def process_temp_frame(temp_frame_path : str, frame_number : int) -> bool:'source_vision_frames': source_vision_frames,'source_audio_frame': source_audio_frame,'source_voice_frame': source_voice_frame,- 'target_vision_frame': target_vision_frame[:, :, :3],+ 'target_vision_frames': target_vision_frames,'temp_vision_frame': temp_vision_frame[:, :, :3],'temp_vision_mask': temp_vision_mask})diff --git a/requirements.txt b/requirements.txtindex ec42df17..e7658235 100644--- a/requirements.txt+++ b/requirements.txt@@ -1,8 +1,8 @@gradio-rangeslider==0.0.8-gradio==5.44.1+gradio==5.50.0numpy==2.2.1onnx==1.21.0-onnxruntime==1.24.4+onnxruntime==1.26.0opencv-python==4.13.0.92tqdm==4.67.3scipy==1.17.1diff --git a/tests/test_audio.py b/tests/test_audio.pyindex 36faf9b0..ca3fb853 100644--- a/tests/test_audio.py+++ b/tests/test_audio.py@@ -17,8 +17,8 @@ def before_all() -> None:def test_get_audio_frame() -> None:- assert hasattr(get_audio_frame(get_test_example_file('source.mp3'), 25), '__array_interface__')- assert hasattr(get_audio_frame(get_test_example_file('source.wav'), 25), '__array_interface__')+ assert get_audio_frame(get_test_example_file('source.mp3'), 25).shape == (80, 16)+ assert get_audio_frame(get_test_example_file('source.wav'), 25).shape == (80, 16)assert get_audio_frame('invalid', 25) is Nonediff --git a/tests/test_common_helper.py b/tests/test_common_helper.pyindex 12e78853..e96a1d8c 100644--- a/tests/test_common_helper.py+++ b/tests/test_common_helper.py@@ -1,4 +1,4 @@-from facefusion.common_helper import calculate_float_step, calculate_int_step, create_float_metavar, create_float_range, create_int_metavar, create_int_range+from facefusion.common_helper import calculate_float_step, calculate_int_step, create_float_metavar, create_float_range, create_int_metavar, create_int_range, get_middledef test_create_int_metavar() -> None:@@ -16,7 +16,7 @@ def test_create_int_range() -> None:def test_create_float_range() -> None:assert create_float_range(0.0, 1.0, 0.5) == [ 0.0, 0.5, 1.0 ]- assert create_float_range(0.0, 1.0, 0.05) == [ 0.0, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50, 0.55, 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90, 0.95, 1.0 ]+ assert create_float_range(0.0, 0.5, 0.05) == [ 0.0, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50 ]def test_calc_int_step() -> None:@@ -25,3 +25,8 @@ def test_calc_int_step() -> None:def test_calc_float_step() -> None:assert calculate_float_step([ 0.1, 0.2 ]) == 0.1+++def test_get_middle() -> None:+ assert get_middle([ 1, 2, 3, 4, 5 ]) == 3+ assert get_middle([ 1 ]) == 1diff --git a/tests/test_config.py b/tests/test_config.pyindex cae75778..ba821508 100644--- a/tests/test_config.py+++ b/tests/test_config.py@@ -1,14 +1,13 @@-from configparser import ConfigParser-import pytest-from facefusion import config+from facefusion import config, state_manager@pytest.fixture(scope = 'module', autouse = True)def before_all() -> None:- config.CONFIG_PARSER = ConfigParser()- config.CONFIG_PARSER.read_dict(+ state_manager.init_item('config_path', 'facefusion.ini')+ config_parser = config.get_static_config_parser()+ config_parser.read_dict({'str':{diff --git a/tests/test_face_creator.py b/tests/test_face_creator.pynew file mode 100644index 00000000..ba59c057--- /dev/null+++ b/tests/test_face_creator.py@@ -0,0 +1,105 @@+import subprocess++import numpy+import pytest++from facefusion import face_classifier, face_detector, face_landmarker, face_recognizer, state_manager+from facefusion.download import conditional_download+from facefusion.face_creator import average_face_geometry, get_many_faces, get_one_face, refill_faces+from facefusion.face_store import clear_faces+from facefusion.vision import read_static_image+from .helper import get_test_example_file, get_test_examples_directory++[email protected](scope = 'module', autouse = True)+def before_all() -> None:+ conditional_download(get_test_examples_directory(),+ [+ 'https://github.com/facefusion/facefusion-assets/releases/download/examples-3.0.0/source.jpg'+ ])+ subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.8:ih*0.8', get_test_example_file('source-80crop.jpg') ])+ subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.7:ih*0.7', get_test_example_file('source-70crop.jpg') ])+ subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.6:ih*0.6', get_test_example_file('source-60crop.jpg') ])++ state_manager.init_item('execution_device_ids', [ 0 ])+ state_manager.init_item('execution_providers', [ 'cpu' ])+ state_manager.init_item('download_providers', [ 'github' ])+ state_manager.init_item('face_detector_angles', [ 0 ])+ state_manager.init_item('face_detector_model', 'many')+ state_manager.init_item('face_detector_size', '640x640')+ state_manager.init_item('face_detector_margin', (0, 0, 0, 0))+ state_manager.init_item('face_detector_score', 0.5)+ state_manager.init_item('face_landmarker_model', 'many')+ state_manager.init_item('face_landmarker_score', 0.5)++ face_classifier.pre_check()+ face_detector.pre_check()+ face_landmarker.pre_check()+ face_recognizer.pre_check()++[email protected](autouse = True)+def before_each() -> None:+ face_classifier.clear_inference_pool()+ face_detector.clear_inference_pool()+ face_landmarker.clear_inference_pool()+ face_recognizer.clear_inference_pool()+ clear_faces()+++def test_get_one_face() -> None:+ source_vision_frame = read_static_image(get_test_example_file('source.jpg'))+ face = get_one_face(get_many_faces([ source_vision_frame ]))++ assert face.bounding_box.size == 4+++def test_get_many_faces() -> None:+ source_path = get_test_example_file('source.jpg')+ source_vision_frame = read_static_image(source_path)+ many_faces = get_many_faces([ source_vision_frame, source_vision_frame, source_vision_frame ])++ assert len(many_faces) == 3+++def test_refill_faces() -> None:+ source_vision_frame = read_static_image(get_test_example_file('source.jpg'))+ face = get_one_face(get_many_faces([ source_vision_frame ]))+ face_first = face._replace(bounding_box = numpy.array([ 0, 0, 10, 10 ]))+ face_middle = face._replace(bounding_box = numpy.array([ 40, 40, 50, 50 ]))+ face_last = face._replace(bounding_box = numpy.array([ 80, 80, 90, 90 ]))++ fill_faces = refill_faces([ face_first, None, face_last ])++ assert fill_faces[0].bounding_box.tolist() == [ 0.0, 0.0, 10.0, 10.0 ]+ assert fill_faces[1].bounding_box.tolist() == [ 40.0, 40.0, 50.0, 50.0 ]+ assert fill_faces[2].bounding_box.tolist() == [ 80.0, 80.0, 90.0, 90.0 ]++ fill_faces = refill_faces([ face_first, None, None, None, face_last ])++ assert fill_faces[0].bounding_box.tolist() == [ 0.0, 0.0, 10.0, 10.0 ]+ assert fill_faces[1].bounding_box.tolist() == [ 20.0, 20.0, 30.0, 30.0 ]+ assert fill_faces[2].bounding_box.tolist() == [ 40.0, 40.0, 50.0, 50.0 ]+ assert fill_faces[3].bounding_box.tolist() == [ 60.0, 60.0, 70.0, 70.0 ]+ assert fill_faces[4].bounding_box.tolist() == [ 80.0, 80.0, 90.0, 90.0 ]++ fill_faces = refill_faces([ face_first, None, face_middle, None, face_last ])++ assert fill_faces[0].bounding_box.tolist() == [ 0.0, 0.0, 10.0, 10.0 ]+ assert fill_faces[1].bounding_box.tolist() == [ 20.0, 20.0, 30.0, 30.0 ]+ assert fill_faces[2].bounding_box.tolist() == [ 40.0, 40.0, 50.0, 50.0 ]+ assert fill_faces[3].bounding_box.tolist() == [ 60.0, 60.0, 70.0, 70.0 ]+ assert fill_faces[4].bounding_box.tolist() == [ 80.0, 80.0, 90.0, 90.0 ]+++def test_average_face_geometry() -> None:+ source_vision_frame = read_static_image(get_test_example_file('source.jpg'))+ face_previous = get_one_face(get_many_faces([ source_vision_frame ]))+ face_next = get_one_face(get_many_faces([ source_vision_frame ]))+ face_previous = face_previous._replace(bounding_box = numpy.array([ 0, 0, 10, 10 ]))+ face_next = face_next._replace(bounding_box = numpy.array([ 80, 80, 90, 90 ]))++ assert average_face_geometry([face_previous, face_next], 0.5).bounding_box.tolist() == [40.0, 40.0, 50.0, 50.0]+ assert average_face_geometry([face_previous, face_next], 0.5).angle == face_next.angle+ assert average_face_geometry([face_previous, face_next], 0.5).embedding is face_next.embedding+ assert average_face_geometry([face_previous, face_next], 0.25).embedding is face_previous.embeddingdiff --git a/tests/test_face_analyser.py b/tests/test_face_detector.pysimilarity index 54%rename from tests/test_face_analyser.pyrename to tests/test_face_detector.pyindex 95d49053..01b0b4c1 100644--- a/tests/test_face_analyser.py+++ b/tests/test_face_detector.py@@ -2,10 +2,10 @@ import subprocessimport pytest-from facefusion import face_classifier, face_detector, face_landmarker, face_recognizer, state_manager+from facefusion import face_detector, state_managerfrom facefusion.download import conditional_download-from facefusion.face_analyser import get_many_faces-from facefusion.face_store import clear_static_faces+from facefusion.face_detector import detect_with_retinaface, detect_with_scrfd, detect_with_yolo_face, detect_with_yunet+from facefusion.face_helper import apply_nms, get_nms_thresholdfrom facefusion.vision import read_static_imagefrom .helper import get_test_example_file, get_test_examples_directory@@ -19,34 +19,23 @@ def before_all() -> None:subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.8:ih*0.8', get_test_example_file('source-80crop.jpg') ])subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.7:ih*0.7', get_test_example_file('source-70crop.jpg') ])subprocess.run([ 'ffmpeg', '-i', get_test_example_file('source.jpg'), '-vf', 'crop=iw*0.6:ih*0.6', get_test_example_file('source-60crop.jpg') ])+state_manager.init_item('execution_device_ids', [ 0 ])state_manager.init_item('execution_providers', [ 'cpu' ])state_manager.init_item('download_providers', [ 'github' ])state_manager.init_item('face_detector_angles', [ 0 ])state_manager.init_item('face_detector_model', 'many')state_manager.init_item('face_detector_score', 0.5)- state_manager.init_item('face_landmarker_model', 'many')- state_manager.init_item('face_landmarker_score', 0.5)- face_classifier.pre_check()- face_landmarker.pre_check()- face_recognizer.pre_check()++ face_detector.pre_check()@pytest.fixture(autouse = True)def before_each() -> None:- face_classifier.clear_inference_pool()face_detector.clear_inference_pool()- face_landmarker.clear_inference_pool()- face_recognizer.clear_inference_pool()- clear_static_faces()-def test_get_one_face_with_retinaface() -> None:- state_manager.init_item('face_detector_model', 'retinaface')- state_manager.init_item('face_detector_size', '320x320')- state_manager.init_item('face_detector_margin', (0, 0, 0, 0))- face_detector.pre_check()-+def test_detect_with_retinaface() -> None:source_paths =\[get_test_example_file('source.jpg'),@@ -57,17 +46,13 @@ def test_get_one_face_with_retinaface() -> None:for source_path in source_paths:source_frame = read_static_image(source_path)- many_faces = get_many_faces([ source_frame ])-- assert len(many_faces) == 1+ bounding_boxes, face_scores, face_landmarks_5 = detect_with_retinaface(source_frame, '320x320')+ keep_indices = apply_nms(bounding_boxes, face_scores, 0.5, get_nms_threshold('retinaface', [ 0 ]))+ assert len(keep_indices) == 1-def test_get_one_face_with_scrfd() -> None:- state_manager.init_item('face_detector_model', 'scrfd')- state_manager.init_item('face_detector_size', '320x320')- state_manager.init_item('face_detector_margin', (0, 0, 0, 0))- face_detector.pre_check()+def test_detect_with_scrfd() -> None:source_paths =\[get_test_example_file('source.jpg'),@@ -78,17 +63,13 @@ def test_get_one_face_with_scrfd() -> None:for source_path in source_paths:source_frame = read_static_image(source_path)- many_faces = get_many_faces([ source_frame ])+ bounding_boxes, face_scores, face_landmarks_5 = detect_with_scrfd(source_frame, '320x320')+ keep_indices = apply_nms(bounding_boxes, face_scores, 0.5, get_nms_threshold('scrfd', [ 0 ]))- assert len(many_faces) == 1+ assert len(keep_indices) == 1-def test_get_one_face_with_yoloface() -> None:- state_manager.init_item('face_detector_model', 'yolo_face')- state_manager.init_item('face_detector_size', '640x640')- state_manager.init_item('face_detector_margin', (0, 0, 0, 0))- face_detector.pre_check()-+def test_detect_with_yolo_face() -> None:source_paths =\[get_test_example_file('source.jpg'),@@ -99,17 +80,13 @@ def test_get_one_face_with_yoloface() -> None:for source_path in source_paths:source_frame = read_static_image(source_path)- many_faces = get_many_faces([ source_frame ])+ bounding_boxes, face_scores, face_landmarks_5 = detect_with_yolo_face(source_frame, '640x640')+ keep_indices = apply_nms(bounding_boxes, face_scores, 0.5, get_nms_threshold('yolo_face', [ 0 ]))- assert len(many_faces) == 1+ assert len(keep_indices) == 1-def test_get_one_face_with_yunet() -> None:- state_manager.init_item('face_detector_model', 'yunet')- state_manager.init_item('face_detector_size', '640x640')- state_manager.init_item('face_detector_margin', (0, 0, 0, 0))- face_detector.pre_check()-+def test_detect_with_yunet() -> None:source_paths =\[get_test_example_file('source.jpg'),@@ -120,14 +97,7 @@ def test_get_one_face_with_yunet() -> None:for source_path in source_paths:source_frame = read_static_image(source_path)- many_faces = get_many_faces([ source_frame ])-- assert len(many_faces) == 1---def test_get_many_faces() -> None:- source_path = get_test_example_file('source.jpg')- source_frame = read_static_image(source_path)- many_faces = get_many_faces([ source_frame, source_frame, source_frame ])+ bounding_boxes, face_scores, face_landmarks_5 = detect_with_yunet(source_frame, '640x640')+ keep_indices = apply_nms(bounding_boxes, face_scores, 0.5, get_nms_threshold('yunet', [ 0 ]))- assert len(many_faces) == 3+ assert len(keep_indices) == 1diff --git a/tests/test_face_tracker.py b/tests/test_face_tracker.pynew file mode 100644index 00000000..66991a5b--- /dev/null+++ b/tests/test_face_tracker.py@@ -0,0 +1,102 @@+import numpy+import pytest++from facefusion import face_classifier, face_detector, face_landmarker, face_recognizer, state_manager+from facefusion.common_helper import get_first, get_last+from facefusion.download import conditional_download+from facefusion.face_creator import get_many_faces, get_one_face+from facefusion.face_store import clear_faces+from facefusion.face_tracker import create_face_tracks, select_face_track, track_faces+from facefusion.vision import read_static_video_chunk, read_static_video_frame+from .helper import get_test_example_file, get_test_examples_directory++[email protected](scope = 'module', autouse = True)+def before_all() -> None:+ conditional_download(get_test_examples_directory(),+ [+ 'https://github.com/facefusion/facefusion-assets/releases/download/examples-3.0.0/target-240p.mp4'+ ])++ state_manager.init_item('execution_device_ids', [ 0 ])+ state_manager.init_item('execution_providers', [ 'cpu' ])+ state_manager.init_item('download_providers', [ 'github' ])+ state_manager.init_item('face_detector_angles', [ 0 ])+ state_manager.init_item('face_detector_model', 'yolo_face')+ state_manager.init_item('face_detector_size', '640x640')+ state_manager.init_item('face_detector_margin', (0, 0, 0, 0))+ state_manager.init_item('face_detector_score', 0.5)+ state_manager.init_item('face_landmarker_model', 'many')+ state_manager.init_item('face_landmarker_score', 0.5)+ state_manager.init_item('face_tracker_score', 0.3)++ face_classifier.pre_check()+ face_detector.pre_check()+ face_landmarker.pre_check()+ face_recognizer.pre_check()++[email protected](autouse = True)+def before_each() -> None:+ face_classifier.clear_inference_pool()+ face_detector.clear_inference_pool()+ face_landmarker.clear_inference_pool()+ face_recognizer.clear_inference_pool()+ clear_faces()+++def test_track_faces() -> None:+ target_path = get_test_example_file('target-240p.mp4')+ video_frame_chunk = read_static_video_chunk(target_path, 0, 7)+ target_vision_frames = [ video_frame_chunk.get(frame_number) for frame_number in sorted(video_frame_chunk) ]+ empty_vision_frame = numpy.zeros_like(get_first(target_vision_frames))++ target_vision_frames[2] = empty_vision_frame+ target_vision_frames[3] = empty_vision_frame+ target_vision_frames[4] = empty_vision_frame+ target_vision_frames[5] = empty_vision_frame++ assert len(track_faces(target_vision_frames, 0.3)) == 1++ target_vision_frames = [ video_frame_chunk.get(frame_number) for frame_number in sorted(video_frame_chunk)[:5] ]+ target_vision_frames[0] = empty_vision_frame+ target_vision_frames[1] = empty_vision_frame+ target_vision_frames[2] = empty_vision_frame++ assert len(track_faces(target_vision_frames, 0.3)) == 0+++def test_create_face_tracks() -> None:+ target_vision_frame = read_static_video_frame(get_test_example_file('target-240p.mp4'), 0)+ multi_face_vision_frame = numpy.hstack([ target_vision_frame, target_vision_frame ])++ face_tracks = create_face_tracks([ target_vision_frame, target_vision_frame ], 0.3)++ assert len(face_tracks) == 1+ assert sorted(get_first(face_tracks)) == [ 0, 1 ]++ face_tracks = create_face_tracks([ multi_face_vision_frame, multi_face_vision_frame ], 0.3)++ assert len(face_tracks) == 2+ assert sorted(get_first(face_tracks)) == [ 0, 1 ]+ assert sorted(get_last(face_tracks)) == [ 0, 1 ]++ assert len(create_face_tracks([ target_vision_frame, target_vision_frame ], 1.0)) == 2+++def test_select_face_track() -> None:+ target_vision_frame = read_static_video_frame(get_test_example_file('target-240p.mp4'), 0)+ face = get_one_face(get_many_faces([ target_vision_frame ]))+ face_overlap = face._replace(bounding_box = numpy.array([ 12, 12, 52, 52 ]))+ face_distant = face._replace(bounding_box = numpy.array([ 200, 200, 240, 240 ]))+ face_track_overlap =\+ {+ 0 : face._replace(bounding_box = numpy.array([ 10, 10, 50, 50 ]))+ }+ face_track_distant =\+ {+ 0 : face._replace(bounding_box = numpy.array([ 100, 100, 140, 140 ]))+ }++ assert select_face_track([ face_track_overlap, face_track_distant ], face_overlap, 0.3) is face_track_overlap+ assert select_face_track([ face_track_overlap, face_track_distant ], face_distant, 0.3) == {}diff --git a/tests/test_ffmpeg.py b/tests/test_ffmpeg.pyindex a3fa0f8f..5fdeca1a 100644--- a/tests/test_ffmpeg.py+++ b/tests/test_ffmpeg.py@@ -9,7 +9,7 @@ from facefusion import process_manager, state_managerfrom facefusion.download import conditional_downloadfrom facefusion.ffmpeg import concat_video, extract_frames, merge_video, read_audio_buffer, replace_audio, restore_audiofrom facefusion.filesystem import copy_file-from facefusion.temp_helper import clear_temp_directory, create_temp_directory, get_temp_file_path, resolve_temp_frame_paths+from facefusion.temp_helper import clear_temp_directory, create_temp_directory, get_temp_file_path, resolve_temp_frame_setfrom facefusion.types import EncoderSetfrom .helper import get_test_example_file, get_test_examples_directory, get_test_output_file, prepare_test_output_directory@@ -85,7 +85,7 @@ def test_extract_frames() -> None:create_temp_directory(target_path)assert extract_frames(target_path, (452, 240), 30.0, trim_frame_start, trim_frame_end) is True- assert len(resolve_temp_frame_paths(target_path)) == frame_total+ assert len(resolve_temp_frame_set(target_path)) == frame_totalclear_temp_directory(target_path)diff --git a/tests/test_ffmpeg_builder.py b/tests/test_ffmpeg_builder.pyindex 1bc9a7c6..2a7b9df1 100644--- a/tests/test_ffmpeg_builder.py+++ b/tests/test_ffmpeg_builder.py@@ -1,7 +1,7 @@from shutil import whichfrom facefusion import ffmpeg_builder-from facefusion.ffmpeg_builder import chain, concat, keep_video_alpha, run, select_frame_range, set_audio_quality, set_audio_sample_size, set_stream_mode, set_video_encoder, set_video_fps, set_video_quality+from facefusion.ffmpeg_builder import chain, concat, keep_video_alpha, run, select_frame_range, set_audio_quality, set_audio_sample_size, set_faststart, set_stream_mode, set_video_encoder, set_video_fps, set_video_quality, set_video_tagdef test_run() -> None:@@ -71,6 +71,24 @@ def test_set_audio_quality() -> None:assert set_audio_quality('flac', 100) == []+def test_set_faststart() -> None:+ assert set_faststart('m4v') == [ '-movflags', '+faststart' ]+ assert set_faststart('mov') == [ '-movflags', '+faststart' ]+ assert set_faststart('mp4') == [ '-movflags', '+faststart' ]+ assert set_faststart('mkv') == []+ assert set_faststart('webm') == []+++def test_set_video_tag() -> None:+ assert set_video_tag('libx265', 'm4v') == [ '-tag:v', 'hvc1' ]+ assert set_video_tag('hevc_nvenc', 'mov') == [ '-tag:v', 'hvc1' ]+ assert set_video_tag('hevc_videotoolbox', 'mp4') == [ '-tag:v', 'hvc1' ]+ assert set_video_tag('libx265', 'mkv') == []+ assert set_video_tag('libx265', 'webm') == []+ assert set_video_tag('libx264', 'mp4') == []+ assert set_video_tag('h264_nvenc', 'mp4') == []++def test_set_video_quality() -> None:assert set_video_quality('libx264', 0) == [ '-crf', '51' ]assert set_video_quality('libx264', 50) == [ '-crf', '26' ]diff --git a/tests/test_memory.py b/tests/test_memory.pydeleted file mode 100644index e637ea10..00000000--- a/tests/test_memory.py+++ /dev/null@@ -1,8 +0,0 @@-from facefusion.common_helper import is_linux, is_macos-from facefusion.memory import limit_system_memory---def test_limit_system_memory() -> None:- assert limit_system_memory(4) is True- if is_linux() or is_macos():- assert limit_system_memory(1024) is Falsediff --git a/tests/test_temp_helper.py b/tests/test_temp_helper.pyindex 6903d2ca..730b7506 100644--- a/tests/test_temp_helper.py+++ b/tests/test_temp_helper.py@@ -5,7 +5,7 @@ import pytestfrom facefusion import state_managerfrom facefusion.download import conditional_download-from facefusion.temp_helper import get_temp_directory_path, get_temp_file_path, get_temp_frames_pattern+from facefusion.temp_helper import get_temp_directory_path, get_temp_file_path, get_temp_frame_patternfrom .helper import get_test_example_file, get_test_examples_directory@@ -29,6 +29,6 @@ def test_get_temp_directory_path() -> None:assert get_temp_directory_path(get_test_example_file('target-240p.mp4')) == os.path.join(temp_directory, 'facefusion', 'target-240p')-def test_get_temp_frames_pattern() -> None:+def test_get_temp_frame_pattern() -> None:temp_directory = tempfile.gettempdir()- assert get_temp_frames_pattern(get_test_example_file('target-240p.mp4'), '%04d') == os.path.join(temp_directory, 'facefusion', 'target-240p', '%04d.png')+ assert get_temp_frame_pattern(get_test_example_file('target-240p.mp4'), '%04d') == os.path.join(temp_directory, 'facefusion', 'target-240p', '%04d.png')diff --git a/tests/test_vision.py b/tests/test_vision.pyindex 35bec382..84bac2d7 100644--- a/tests/test_vision.py+++ b/tests/test_vision.py@@ -1,11 +1,12 @@import osimport subprocess+import numpyimport pytestfrom facefusion.common_helper import is_linuxfrom facefusion.download import conditional_download-from facefusion.vision import calculate_histogram_difference, count_trim_frame_total, count_video_frame_total, detect_image_resolution, detect_video_duration, detect_video_fps, detect_video_resolution, match_frame_color, normalize_resolution, pack_resolution, predict_video_frame_total, read_image, read_video_frame, restrict_image_resolution, restrict_trim_frame, restrict_video_fps, restrict_video_resolution, scale_resolution, unpack_resolution, write_image+from facefusion.vision import calculate_histogram_difference, count_trim_frame_total, count_video_frame_total, detect_image_resolution, detect_video_duration, detect_video_fps, detect_video_resolution, match_frame_color, normalize_resolution, pack_resolution, predict_video_frame_total, read_image, read_video_chunk, read_video_frame, restrict_image_resolution, restrict_trim_frame, restrict_video_fps, restrict_video_resolution, scale_resolution, select_video_frames, unpack_resolution, write_imagefrom .helper import get_test_example_file, get_test_examples_directory, get_test_output_file, prepare_test_output_directory@@ -63,10 +64,27 @@ def test_restrict_image_resolution() -> None:def test_read_video_frame() -> None:- assert hasattr(read_video_frame(get_test_example_file('target-240p-25fps.mp4')), '__array_interface__')+ target_path = get_test_example_file('target-240p-25fps.mp4')++ assert read_video_frame(target_path).shape == (226, 426, 3)+ assert numpy.array_equal(read_video_frame(target_path, 49), select_video_frames(target_path, 49, 5)[5])+ assert numpy.array_equal(read_video_frame(target_path, 50), select_video_frames(target_path, 50, 5)[5])+ assert numpy.array_equal(read_video_frame(target_path, 51), select_video_frames(target_path, 51, 5)[5])assert read_video_frame('invalid') is None+def test_read_video_chunk() -> None:+ assert len(read_video_chunk(get_test_example_file('target-240p-25fps.mp4'), 1, 40)) == 40+ assert read_video_chunk('invalid', 1, 40) == {}+++def test_select_video_frames() -> None:+ assert len(select_video_frames(get_test_example_file('target-240p-25fps.mp4'), 50, 5)) == 11+ assert len(select_video_frames(get_test_example_file('target-240p-25fps.mp4'), 1, 5)) == 11+ assert len(select_video_frames(get_test_example_file('target-240p-25fps.mp4'), 269, 5)) == 11+ assert select_video_frames('invalid', 50, 5) == []++def test_count_video_frame_total() -> None:assert count_video_frame_total(get_test_example_file('target-240p-25fps.mp4')) == 270assert count_video_frame_total(get_test_example_file('target-240p-30fps.mp4')) == 324