VideoMaMa：基于生成先验的掩码引导视频抠图
VideoMaMa: Mask-Guided Video Matting via Generative Prior
利用预训练视频扩散模型的生成先验，将粗分割掩码细化为像素级 alpha matte，并构建大规模伪标签视频抠图数据集，为视频抠图提供可扩展的监督信号。
原文依据 converts coarse segmentation masks into pixel accurate alpha mattes, by leveraging pretrained video diffusion models
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论由于标注数据稀缺，将视频抠图模型泛化到真实世界视频仍是一项重大挑战。为此，我们提出视频掩码到抠图模型 VideoMaMa，利用预训练视频扩散模型将粗分割掩码转换为像素级精确的 alpha matte。尽管仅使用合成数据训练，VideoMaMa 对真实世界视频展现出强大的零样本泛化能力。基于此能力，我们开发了面向大规模视频抠图的可扩展伪标签流程，并构建了 Matting Anything in Video MA-V 数据集，为超过 5 万段涵盖多样场景与运动的真实世界视频提供高质量抠图标注。为验证该数据集的有效性，我们在 MAV 上微调 SAM2 模型得到 SAM2-Matte，其在真实场景视频上的鲁棒性优于在现有抠图数据集上训练的同一模型。这些发现强调了大规模伪标签视频抠图的重要性，并展示了生成先验与易获取的分割线索如何推动视频抠图研究的可扩展进展。
英文原文摘要 1149 字符
Generalizing video matting models to real-world videos remains a significant challenge due to the scarcity of labeled data. To address this, we present Video Mask-to-Matte Model VideoMaMa that converts coarse segmentation masks into pixel accurate alpha mattes, by leveraging pretrained video diffusion models. VideoMaMa demonstrates strong zero-shot generalization to real-world footage, even though it is trained solely on synthetic data. Building on this capability, we develop a scalable pseudo-labeling pipeline for large-scale video matting and construct the Matting Anything in Video MA-V dataset, which offers high-quality matting annotations for more than 50K real-world videos spanning diverse scenes and motions. To validate the effectiveness of this dataset, we fine-tune the SAM2 model on MAV to obtain SAM2-Matte, which outperforms the same model trained on existing matting datasets in terms of robustness on in-the-wild videos.These findings emphasize the importance of large-scale pseudo-labeled video matting and showcase how generative priors and accessible segmentation cues can drive scalable progress in video matting research.