File size: 1,535 Bytes
a2a5d5f | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 | ---
license: other
license_name: s-lab-1.0
license_link: https://huggingface.co/AEmotionStudio/matanyone2/blob/main/LICENSE.txt
tags:
- video-matting
- alpha-matting
- comfyui
- ffmpega
---
# MatAnyone2 — SafeTensors Mirror
This is a **SafeTensors conversion** of the [MatAnyone2](https://github.com/pq-yang/MatAnyone2) video matting model (CVPR 2026) for use with [ComfyUI-FFMPEGA](https://github.com/AEmotionStudio/ComfyUI-FFMPEGA).
## Model Details
| Property | Value |
|----------|-------|
| Architecture | MatAnyone2 (temporal video matting) |
| Parameters | 562 tensors, ~135 MB |
| Format | SafeTensors |
| Original Format | PyTorch (.pth) |
| Source | [pq-yang/MatAnyone2](https://github.com/pq-yang/MatAnyone2) |
## ⚠️ License Notice
This model is released under the **NTU S-Lab License 1.0** which is strictly **non-commercial**.
> Redistribution and use for non-commercial purpose in source and binary forms, with or without modification, are permitted provided that the following conditions are met...
See [LICENSE.txt](LICENSE.txt) for the full license text. **Commercial use requires explicit written permission from the original authors.**
## Usage
This model is automatically downloaded by ComfyUI-FFMPEGA when the `video_matting` no-LLM mode is selected. No manual setup is required.
## Citation
```bibtex
@article{yang2025matanyone2,
title={MatAnyone2: Stable Video Matting with Consistent Memory Propagation},
author={Yang, Peiqing and Shang, Shangchen and others},
year={2025}
}
```
|