T
ToolShelf
VIDEOMAMA
// Official implementation of "VideoMaMa: Mask-Guided Video Matting via Generative Prior", CVPR 2026

VideoMaMa

Official implementation of "VideoMaMa: Mask-Guided Video Matting via Generative Prior", CVPR 2026

13EmergingUnknown
License
NOASSERTION
Updated
Today

What it does

VideoMaMa: Mask-Guided Video Matting via Generative Prior Sangbeom Lim1 · Seoung Wug Oh2 · Jiahui Huang2 · Heeji Yoon3 Seungryong Kim3 · Joon-Young Lee2 1Korea University    2Adobe Research    3KAIST AI ArXiv 2026 VideoMaMa is a mask-guided video matting framework that leverages a video generative prior. By utilizing this prior, it supports stable

Getting Started

git
git clone https://github.com/cvlab-kaist/VideoMaMa

Platforms

🪟windows🍎mac🐧linux

Install Difficulty

moderate

Built With

python

Community Reactions