Robuta

https://videomatting.com/ Video Matting Benchmark VideoMatting project is the first public objective benchmark of video matting methods. It is based on a high-quality ground truth alpha sequences collected... video mattingbenchmark https://www.tovideo.org/models/matanyone MatAnyone: Advanced Video Matting for Stable and Precise Human Segmentation Discover MatAnyone, a robust AI framework for video matting. Achieve stable target tracking, fine boundary detail extraction, and adaptive video segmentation... advanced videomatanyonemattingstableprecise https://arxiv.org/abs/2512.11782 [2512.11782] MatAnyone 2: Scaling Video Matting via a Learned Quality Evaluator Abstract page for arXiv paper 2512.11782: MatAnyone 2: Scaling Video Matting via a Learned Quality Evaluator video mattingmatanyonescalingvialearned https://studio.aifilms.ai/blog/matanyone-2-video-matting MatAnyone 2: AI Video Matting Accepted to CVPR 2026 Mar 12, 2026 - MatAnyone 2 extracts foreground subjects with precise alpha mattes, accepted to CVPR 2026 with a 28K clip real world dataset and open source code. ai videomatanyonemattingacceptedcvpr https://replicate.com/arielreplicate/robust_video_matting arielreplicate/robust_video_matting | Run with an API on Replicate extract foreground of a video video mattingrobustrunapireplicate