提出解耦记忆选择策略，提升多目标视频分割稳定性
SAM3-DMS: Decoupled Memory Selection for Multi-target Video Segmentation of SAM3
- 对每个目标独立进行记忆选择，避免集体决策偏差
- 在高密度目标场景下表现更优，保持目标身份一致性
- 无需训练，可直接集成到SAM3中，适合真实复杂场景
Segment Anything 3 (SAM3) 已建立强大的基础，能稳健地检测、分割并跟踪视频中的指定目标。然而，在原始实现中，其基于组级别的集体记忆选择在复杂多目标场景下表现不佳，因其对所有并发目标采用同步决策，且依赖平均性能，常忽略单个目标的可靠性。为此，我们提出 SAM3-DMS，一种无需训练的解耦策略，对每个目标进行细粒度的记忆选择。实验表明，该方法实现了稳健的身份保持和跟踪稳定性。尤其在目标密度增加时，优势更加明显，为真实世界中同时进行多目标视频分割奠定了坚实基础。
原文摘要 · Abstract (English)
Segment Anything 3 (SAM3) has established a powerful foundation that robustly detects, segments, and tracks specified targets in videos. However, in its original implementation, its group-level collective memory selection is suboptimal for complex multi-object scenarios, as it employs a synchronized decision across all concurrent targets conditioned on their average performance, often overlooking individual reliability. To this end, we propose SAM3-DMS, a training-free decoupled strategy that utilizes fine-grained memory selection on individual objects. Experiments demonstrate that our approach achieves robust identity preservation and tracking stability. Notably, our advantage becomes more pronounced with increased target density, establishing a solid foundation for simultaneous multi-target video segmentation in the wild.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。