ARC-SORT: Adaptive, Robust, Compensated Multi-Object Tracking with Interacting Multiple Models

📄 arXiv: 2610.22306v1 📥 PDF

作者: Girish Kumar Adari

分类: cs.CV

发布日期: 2026-10-15

备注: 14 pages, 4 figures, 13 tables; supplementary material included as appendix; qualitative video as ancillary file

💡 一句话要点

提出ARC-SORT以解决体育赛事中的多目标跟踪问题

🎯 匹配领域: 支柱七：动作重定向 (Motion Retargeting) 支柱八：物理动画 (Physics-based Animation)

关键词: 多目标跟踪 卡尔曼滤波 自适应算法 实时处理 体育赛事分析

📋 核心要点

  1. 现有多目标跟踪方法在快速移动和方向变化的运动员跟踪中表现不佳，尤其是在摄像机动态变化时。
  2. ARC-SORT通过自适应不确定性测量和交互式多模型卡尔曼滤波器，动态调整搜索区域，提高跟踪精度。
  3. 在SportsMOT和DanceTrack测试集上，ARC-SORT分别达到了81.82和62.1的HOTA，且身份切换次数最少，表现优异。

📝 摘要（中文）

在体育赛事转播中，多目标跟踪面临运动员快速移动和摄像机同时平移缩放的挑战。本文提出了ARC-SORT（自适应、鲁棒、补偿的SORT），一种基于检测的跟踪算法，具有实时运行能力且无需训练。该算法通过测量自身的不确定性来决定行动，采用交互式多模型（IMM）卡尔曼滤波器为每个轨迹提供预测协方差，并基于此生成每个轨迹的搜索区域。实验结果表明，ARC-SORT在SportsMOT和DanceTrack测试集上均显著提升了跟踪精度，且在处理离开帧的运动员时，通过外观重新识别有效链接轨迹。ARC-SORT在SportsMOT测试集上达到了81.82的HOTA，且在69个排行榜条目中身份切换最少。

🔬 方法详解

问题定义：本文旨在解决体育赛事中快速移动运动员的多目标跟踪问题，现有方法在摄像机动态变化时难以保持准确性和鲁棒性。

核心思路：ARC-SORT通过测量自身的不确定性来动态调整搜索区域，采用IMM卡尔曼滤波器为每个轨迹提供预测协方差，从而提高跟踪的准确性和适应性。

技术框架：ARC-SORT的整体架构包括检测阶段、轨迹预测和更新阶段，以及身份纠正阶段。首先进行目标检测，然后利用卡尔曼滤波器进行轨迹预测，最后在轨迹交叉时进行身份纠正。

关键创新：ARC-SORT的主要创新在于引入了不确定性自适应扩展IoU（EIoU），使得在运动预测不可靠时能够扩大搜索区域，从而提高匹配成功率。

关键设计：在设计中，使用了IMM卡尔曼滤波器来计算每个轨迹的协方差，并根据协方差动态调整搜索区域。此外，身份纠正机制在轨迹交叉后进行，确保只有在有足够证据时才进行身份交换。

🖼️ 关键图片

fig_0
img_1
img_2

📊 实验亮点

ARC-SORT在SportsMOT测试集上达到了81.82的HOTA，身份切换次数在69个排行榜条目中最少，同时在DanceTrack测试集上也取得了62.1的HOTA，展现了显著的跟踪精度提升。与Deep-EIoU相比，ARC-SORT在公共检测和嵌入上提升了3.1的HOTA，表现出色。

🎯 应用场景

ARC-SORT在体育赛事转播、监控视频分析和自动驾驶等领域具有广泛的应用潜力。其实时跟踪能力和高精度特性使其能够在复杂动态环境中有效识别和跟踪多个目标，提升了相关领域的智能化水平和应用价值。

📄 摘要（原文）

Multi-object tracking in sports broadcasts is hard for two reasons: athletes move fast and change direction abruptly, and the camera pans and zooms at the same time. We present ARC-SORT (Adaptive, Robust, Compensated SORT), a tracking-by-detection tracker that trains nothing and runs in real time. Its design rule: the tracker measures its own uncertainty, and that measurement decides how it acts. An Interacting Multiple Model (IMM) Kalman filter gives every track a predictive covariance, and we turn it into a per-track search region: the less reliable the motion prediction, the larger the region in which a detection may match (uncertainty-adaptive Expansion IoU, EIoU). On identical inputs, this continuous coupling is what matters: the same covariance used as a hard Mahalanobis gate hurts at every threshold we swept, and richer covariance summaries add nothing measurable. When two tracks cross, the assignment cost itself reports whether the decision was close. ARC-SORT keeps such pairs under review and applies at most one corrective identity swap once post-crossing appearance evidence shows the first decision was wrong; this adds +1.0 HOTA (higher order tracking accuracy) on both the SportsMOT and DanceTrack test sets at under a millisecond per frame. For players who leave the frame for seconds, motion extrapolation carries no information, so our offline re-identification drops the usual motion gate and links tracklets by appearance alone. ARC-SORT reaches 81.82 HOTA on the SportsMOT test set with the fewest identity switches of the 69 leaderboard entries as of August 2026 and tied-third multi-object tracking accuracy (MOTA, 97.09), and 62.1 HOTA on DanceTrack test with the same association settings. On Deep-EIoU's public detections and embeddings it reaches 80.31 HOTA, +3.1 over Deep-EIoU. It runs at about 50 frames per second on a laptop RTX 3060 GPU. Code will be released.