用分割驱动追踪，突破传统检测关联范式，零样本实现顶尖跟踪效果。
SAM2MOT: A Novel Paradigm of Multi-Object Tracking by Segmentation
- 以分割为核心构建追踪框架，直接解决误检和遮挡问题。
- 在DanceTrack上达成+2.1 HOTA和+4.5 IDF1的提升，领先现有方法。
- 无需微调，利用预训练模型实现零样本跨数据集追踪，适合部署应用。
受Segment Anything 2启发，该文提出SAM2MOT——一种基于分割的多目标追踪新范式，摆脱传统检测-关联框架。与以往将分割视为辅助信息不同，SAM2MOT将其置于追踪核心，系统应对误检和遮挡等挑战。在DanceTrack、UAVDT和BDD100K等主流MOT基准上验证有效性。此外，SAM2MOT将预训练检测器、分割器与追踪逻辑融合，构建无需微调的零样本追踪系统，大幅降低对标注数据依赖，推动追踪研究从任务专用转向通用系统。实验显示其在DanceTrack上相较现有方法提升+2.1 HOTA和+4.5 IDF1，达到当前最优水平。代码已开源：https://github.com/TripleJoy/SAM2MOT。
原文摘要 · Abstract (English)
Inspired by Segment Anything 2, which generalizes segmentation from images to videos, we propose SAM2MOT--a novel segmentation-driven paradigm for multi-object tracking that breaks away from the conventional detection-association framework. In contrast to previous approaches that treat segmentation as auxiliary information, SAM2MOT places it at the heart of the tracking process, systematically tackling challenges like false positives and occlusions. Its effectiveness has been thoroughly validated on major MOT benchmarks. Furthermore, SAM2MOT integrates pre-trained detector, pre-trained segmentor with tracking logic into a zero-shot MOT system that requires no fine-tuning. This significantly reduces dependence on labeled data and paves the way for transitioning MOT research from task-specific solutions to general-purpose systems. Experiments on DanceTrack, UAVDT, and BDD100K show state-of-the-art results. Notably, SAM2MOT outperforms existing methods on DanceTrack by +2.1 HOTA and +4.5 IDF1, highlighting its effectiveness in MOT. Code is available at https://github.com/TripleJoy/SAM2MOT.
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。