{"ID":2840775,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.13715","arxiv_id":"2511.13715","title":"Segment Anything Across Shots: A Method and Benchmark","abstract":"This work focuses on multi-shot semi-supervised video object segmentation (MVOS), which aims at segmenting the target object indicated by an initial mask throughout a video with multiple shots. The existing VOS methods mainly focus on single-shot videos and struggle with shot discontinuities, thereby limiting their real-world applicability. We propose a transition mimicking data augmentation strategy (TMA) which enables cross-shot generalization with single-shot data to alleviate the severe annotated multi-shot data sparsity, and the Segment Anything Across Shots (SAAS) model, which can detect and comprehend shot transitions effectively. To support evaluation and future study in MVOS, we introduce Cut-VOS, a new MVOS benchmark with dense mask annotations, diverse object categories, and high-frequency transitions. Extensive experiments on YouMVOS and Cut-VOS demonstrate that the proposed SAAS achieves state-of-the-art performance by effectively mimicking, understanding, and segmenting across complex transitions. The code and datasets are released at https://henghuiding.com/SAAS/.","short_abstract":"This work focuses on multi-shot semi-supervised video object segmentation (MVOS), which aims at segmenting the target object indicated by an initial mask throughout a video with multiple shots. The existing VOS methods mainly focus on single-shot videos and struggle with shot discontinuities, thereby limiting their rea...","url_abs":"https://arxiv.org/abs/2511.13715","url_pdf":"https://arxiv.org/pdf/2511.13715v1","authors":"[\"Hengrui Hu\",\"Kaining Ying\",\"Henghui Ding\"]","published":"2025-11-17T18:58:40Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","project_urls":"[\"https://henghuiding.com/SAAS/\"]","has_code":false,"code_links":[{"ID":607000,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2840775,"paper_url":"https://arxiv.org/abs/2511.13715","paper_title":"Segment Anything Across Shots: A Method and Benchmark","repo_url":"https://github.com/FudanCVL/SAAS","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}