SportsTrack: An Innovative Method for Tracking Athletes in Sports Scenes
This work addresses tracking challenges in sports analytics, but it is incremental as it builds on existing tracking-by-detection methods with specific innovations for sports scenes.
The paper tackled the problem of multiple object tracking of athletes in sports scenes, achieving state-of-the-art results on the SportsMOT dataset by introducing a three-stage matching process and one-to-many correspondence to handle motion blur and body overlapping.
The SportsMOT dataset aims to solve multiple object tracking of athletes in different sports scenes such as basketball or soccer. The dataset is challenging because of the unstable camera view, athletes' complex trajectory, and complicated background. Previous MOT methods can not match enough high-quality tracks of athletes. To pursue higher performance of MOT in sports scenes, we introduce an innovative tracker named SportsTrack, we utilize tracking by detection as our detection paradigm. Then we will introduce a three-stage matching process to solve the motion blur and body overlapping in sports scenes. Meanwhile, we present another innovation point: one-to-many correspondence between detection bboxes and crowded tracks to handle the overlap of athletes' bodies during sports competitions. Compared to other trackers such as BOT-SORT and ByteTrack, We carefully restored edge-lost tracks that were ignored by other trackers. Finally, we reached the SOTA result in the SportsMOT dataset.