Skip to content

[CVPR 2026] PersonaLive! : Expressive Portrait Image Animation for Live Streaming

update readme

Zhiyuan Li3640d9691297
diff --git a/README.md b/README.md
index 6560af2..5eedd1e 100644
--- a/README.md
+++ b/README.md
@@ -13,7 +13,7 @@
</div>
## ๐Ÿ“ฃ Updates
-- **[2025.12.10]** ๐Ÿ”ฅ Release `inference code`, `config` and `checkpoints`!
+- **[2025.12.13]** ๐Ÿ”ฅ Release `inference code`, `config` and `pretrained weights`!
## โš™๏ธ Framework
<img src="assets/overview.png" alt="Image 1" width="100%">
@@ -72,12 +72,12 @@ pretrained_weights
โ””โ”€โ”€ unet_work.engine
```
-### ๐ŸŽฎ Inference (Offline)
+### ๐ŸŽž๏ธ Offline Inference
```
python inference_offline.py
```
-### ๐Ÿ’ป Inference (Online)
-#### Installation
+### ๐Ÿ“ธ Online Inference
+#### ๐Ÿ“ฆ Setup Web UI
```
# install Node.js 18+
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.1/install.sh | bash
@@ -87,13 +87,13 @@ cd webcam
source start.sh
```
-#### TensorRT (Optional)
-Converting the model to TensorRT can significantly speed up inference (~ 2x). Building the engine may take about `20 minutes` depending on your device. Note that TensorRT optimizations may lead to slight variations or a small drop in output quality.
+#### ๐ŸŽ๏ธ Acceleration (Optional)
+Converting the model to TensorRT can significantly speed up inference (~ 2x โšก๏ธ). Building the engine may take about `20 minutes` depending on your device. Note that TensorRT optimizations may lead to slight variations or a small drop in output quality.
```
python torch2trt.py
```
-#### Run
+#### โ–ถ๏ธ Start Streaming
```
python inference_online.py
```
@@ -105,6 +105,5 @@ If you find PersonaLive useful for your research, welcome to ๐ŸŒŸ this repo and
``` -->
-
## โค๏ธ Acknowledgement
This project is based on [Moore-AnimateAnyone](https://github.com/MooreThreads/Moore-AnimateAnyone), [X-NeMo](https://byteaigc.github.io/X-Portrait2/), [StreamDiffusion](https://github.com/cumulo-autumn/StreamDiffusion), [RAIN](https://pscgylotti.github.io/pages/RAIN/) and [LivePortrait](https://github.com/KlingTeam/LivePortrait), thanks to their invaluable contributions.