遮挡感知 SORT：面向鲁棒多目标跟踪的遮挡观测
Occlusion-Aware SORT: Observing Occlusion for Robust Multi-Object Tracking
OA-SORT 把遮挡状态显式建模成可复用的代价修正项（OAO）与动量平滑（BAM），且无需训练即可插入 SORT 类跟踪器；做 MOT 数据关联时可直接借用其高斯图遮挡估计来缓解位置代价混淆，并作为提升现有跟踪器 HOTA/IDF1 的通用后处理。
原文依据 a plug-and-play and training-free framework that includes the Occlusion-Aware Module (OAM), the Occlusion-Aware Offset (OAO), and the Bias-Aware Momentum (BAM)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多目标跟踪（MOT）涉及分析视频序列中的目标轨迹并统计目标数量。然而，由于部分遮挡引起的位置代价混淆，2D MOT 面临诸多挑战。为解决该问题，本文提出新颖的遮挡感知 SORT（OA-SORT）框架，这是一种即插即用且无需训练的框架，包含遮挡感知模块（OAM）、遮挡感知偏移（OAO）与偏置感知动量（BAM）。具体而言，OAM 分析目标的遮挡状态，并引入高斯图（GM）以降低背景影响；而 OAO 与 BAM 则利用 OAM 所描述的遮挡状态来缓解代价混淆并抑制估计的不稳定性。在 DanceTrack、SportsMOT 与 MOT17 数据集上的全面评估表明，遮挡处理在 MOT 中具有重要意义。在 DanceTrack 测试集上，OA-SORT 在 HOTA 与 IDF1 上分别达到 63.1% 与 64.2%。此外，将遮挡感知框架集成到另外四个跟踪器中，可使 HOTA 与 IDF1 平均分别提升 2.08% 与 3.05%，证明了遮挡感知的可复用性。
英文原文摘要 1160 字符
Multi-object tracking (MOT) involves analyzing object trajectories and counting the number of objects in video sequences. However, 2D MOT faces challenges due to positional cost confusion arising from partial occlusion. To address this issue, we present the novel Occlusion-Aware SORT (OA-SORT) framework, a plug-and-play and training-free framework that includes the Occlusion-Aware Module (OAM), the Occlusion-Aware Offset (OAO), and the Bias-Aware Momentum (BAM). Specifically, OAM analyzes the occlusion status of objects, where a Gaussian Map (GM) is introduced to reduce background influence. In contrast, OAO and BAM leverage the OAM-described occlusion status to mitigate cost confusion and suppress estimation instability. Comprehensive evaluations on the DanceTrack, SportsMOT, and MOT17 datasets demonstrate the importance of occlusion handling in MOT. On the DanceTrack test set, OA-SORT achieves 63.1% and 64.2% in HOTA and IDF1, respectively. Furthermore, integrating the Occlusion-Aware framework into the four additional trackers improves HOTA and IDF1 by an average of 2.08% and 3.05%, demonstrating the reusability of the occlusion awareness.