Image Pipeline Playground
"""Shared helpers for OpenCV node implementations."""
from __future__ import annotations
from typing import Any
import numpy as np
from app.models.graph import ParamField, PortDirection, PortSpec
def image_in(
port_id: str = "image",
name: str = "Image",
*,
optional: bool = False,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.INPUT,
data_type="image",
optional=optional,
)
def image_out(
port_id: str = "image",
name: str = "Image",
*,
multiple: bool = False,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.OUTPUT,
data_type="image",
multiple=multiple,
)
def mask_in(
port_id: str = "mask",
name: str = "Mask",
*,
optional: bool = True,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.INPUT,
data_type="mask",
optional=optional,
)
def mask_out(
port_id: str = "mask",
name: str = "Mask",
*,
multiple: bool = False,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.OUTPUT,
data_type="mask",
multiple=multiple,
)
def bboxes_in(
port_id: str = "bboxes",
name: str = "BBoxes",
*,
optional: bool = True,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.INPUT,
data_type="bboxes",
optional=optional,
)
def bboxes_out(
port_id: str = "bboxes",
name: str = "BBoxes",
*,
multiple: bool = False,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.OUTPUT,
data_type="bboxes",
multiple=multiple,
)
def keypoints_in(
port_id: str = "keypoints",
name: str = "Keypoints",
*,
optional: bool = True,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.INPUT,
data_type="keypoints",
optional=optional,
)
def keypoints_out(
port_id: str = "keypoints",
name: str = "Keypoints",
*,
multiple: bool = False,
) -> PortSpec:
return PortSpec(
id=port_id,
name=name,
direction=PortDirection.OUTPUT,
data_type="keypoints",
multiple=multiple,
)
ANNOTATION_DATA_TYPES = frozenset({"bboxes", "keypoints"})
IMAGE_LIKE_DATA_TYPES = frozenset({"image", "mask"})
def require_image(
inputs: dict[str, np.ndarray | list[np.ndarray] | None],
port: str = "image",
) -> np.ndarray:
value = inputs.get(port)
if value is None:
raise ValueError(f"Missing required image input '{port}'")
if isinstance(value, list):
if not value:
raise ValueError(f"Empty image list for input '{port}'")
return value[0]
return value
def number_param(
name: str,
label: str,
default: float,
minimum: float | None = None,
maximum: float | None = None,
step: float | None = None,
description: str | None = None,
) -> ParamField:
return ParamField(
name=name,
label=label,
type="number",
default=default,
minimum=minimum,
maximum=maximum,
step=step,
description=description,
)
def int_param(
name: str,
label: str,
default: int,
minimum: int | None = None,
maximum: int | None = None,
description: str | None = None,
) -> ParamField:
return ParamField(
name=name,
label=label,
type="integer",
default=default,
minimum=minimum,
maximum=maximum,
step=1,
description=description,
)
def select_param(
name: str,
label: str,
default: str,
options: list[str],
description: str | None = None,
) -> ParamField:
return ParamField(
name=name,
label=label,
type="select",
default=default,
options=options,
description=description,
)
def string_param(
name: str,
label: str,
default: str,
description: str | None = None,
) -> ParamField:
return ParamField(
name=name,
label=label,
type="string",
default=default,
description=description,
)
IMAGE_EXTENSIONS = [".png", ".jpg", ".jpeg", ".bmp", ".tif", ".tiff", ".webp", ".gif"]
def file_param(
name: str,
label: str,
default: str = "",
accept: list[str] | None = None,
description: str | None = None,
) -> ParamField:
return ParamField(
name=name,
label=label,
type="file",
default=default,
accept=accept or list(IMAGE_EXTENSIONS),
description=description,
)
def format_params(params: dict[str, Any]) -> str:
items = ", ".join(f"{key}={value!r}" for key, value in sorted(params.items()))
return items