Research map / Video / Video segmentation and tracking
Matting Anything 2: Towards Video Matting for Anything
Chenyi Zhang, Yiheng Lin, Yunchao Wei, Hongsong Wang, Caifeng Shan, Fang Zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Promptable Dual-mode Decoder instead of parallel trimap branch (ablation) | MAM2 · NOVM | MAD: 26.19 -> 18.55 | Improved |
| use Promptable Dual-mode Decoder instead of parallel trimap branch (ablation) | MAM2 · YoutubeMatte | MAD: 1.54 -> 1.16 | Improved |
| add Memory-Separable Siamese second pass on memory-free features (ablation) | MAM2 with PDD · NOVM | MAD: 18.55 -> 14.72 | Improved |
| add Memory-Separable Siamese second pass (ablation) | MAM2 with PDD · YoutubeMatte | MAD: 1.16 -> 1.16 | No clear effect |
| second PDD pass on memory-embedded features (MCS) (ablation) | MAM2 with PDD · NOVM | MAD: 18.55 -> 20.23 | Hurt |
| train MAM2 on the smaller FTP-VM training set (ablation) | MAM2 · NOVM | MAD: 14.72 -> 18.01 | Hurt |
| train MAM2 on the smaller FTP-VM training set (ablation) | MAM2 · YoutubeMatte | MAD: 1.16 -> 1.08 | Improved |
| drive MAM2 with box and point prompts instead of mask | MAM2 · NOVM | MAD: 15.19 -> 14.72 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MatAnyone — MatAnyone: Stable Video Matting with Consistent Memory Propagation (CVPR 2025) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- VideoMaMa: Mask-Guided Video Matting via Generative Prior (CVPR 2026 · similarity 0.92)
- Adaptive Human Matting for Dynamic Videos (CVPR 2023 · similarity 0.91)
- SAM2Matting: Generalized Image and Video Matting (ECCV 2026 · similarity 0.90)
- MatAnyone: Stable Video Matting with Consistent Memory Propagation (CVPR 2025 · similarity 0.90)
- Unifying Automatic and Interactive Matting with Pretrained ViTs (CVPR 2024 · similarity 0.89)
- MaGGIe: Masked Guided Gradual Human Instance Matting (CVPR 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
