Adaptive multi-object tracking improves accuracy in fast sports videos
ARC-SORT: Adaptive, Robust, Compensated Multi-Object Tracking with Interacting Multiple Models
Computer Vision and Pattern Recognition
Summary
Tracking multiple players in sports videos is hard because athletes move quickly and cameras zoom or pan at the same time. The authors introduce ARC-SORT, a tracking method that estimates how uncertain it is about each player’s movement and uses that to decide where to look for matches in the next frame. If players cross paths and cause confusion, ARC-SORT carefully rechecks and fixes identity mistakes using appearance clues after the event. This approach runs fast without extra training and achieves high accuracy on sports and dance videos.
What this means in practice
- •For sports analytics developers: Enhance athlete tracking accuracy and reduce identity errors in broadcast sports video analysis using a model that adapts its search based on motion uncertainty and appearance correction.
- •For surveillance system engineers: Improve multi-person tracking robustness in complex scenes with abrupt movements and camera changes by integrating uncertainty-adaptive search regions and post-event identity correction.
Authors
Girish Kumar Adari
Abstract
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.