Improve inference usage

README.md

@@ -12,19 +12,20 @@ This repository contains the model described in [MatAnyone: Stable Video Matting
 Code: https://github.com/pq-yang/MatAnyone
 
 ## Usage
-Requires `pip install huggingface_hub`.
+Models can be directly loaded from [Hugging Face](https://huggingface.co/PeiqingYang/MatAnyone) to make inference.
 
 ```shell
 pip install -q git+https://github.com/pq-yang/MatAnyone
 ```
 
-Users will only need to use these two lines to load and instantiate the model.
+To extract the foreground and the alpha video you can **directly run the following lines**. Please refer to [inference_hf.py](https://github.com/pq-yang/MatAnyone/blob/main/inference_hf.py) for more arguments.
 ```python
-from matanyone.model.matanyone import MatAnyone
-matanyone = MatAnyone.from_pretrained("PeiqingYang/MatAnyone").cuda().eval()
-```
+from matanyone import InferenceCore
+processor = InferenceCore("PeiqingYang/MatAnyone")
 
-We also provide a **stand-alone** [inference script](https://huggingface.co/PeiqingYang/MatAnyone/blob/main/inference_matanyone_api.py) that could be used as **API**.
-```shell
-python inference_matanyone_api.py -i <input video> -m <first-frame seg mask>
+foreground_path, alpha_path = processor.process_video(
+    input_path = <video_path>,
+    mask_path = <mask_path>,
+    output_path = <output_path>
+)
 ```