No Train Yet Gain: Towards Generic Multi-Object Tracking in Sports and Beyond

TL;DR

This paper introduces McByte, a training-free multi-object tracking framework that leverages mask propagation for robust and adaptable tracking in sports and general scenarios, outperforming traditional methods.

Contribution

McByte is a novel, training-free tracking-by-detection approach that uses temporally propagated segmentation masks to enhance robustness and generalizability across diverse tracking tasks.

Findings

McByte achieves strong performance on SportsMOT, DanceTrack, SoccerNet, and MOT17 datasets.

Mask propagation improves tracking robustness without the need for per-video tuning.

The method works effectively across sports and pedestrian tracking scenarios.

Abstract

Multi-object tracking (MOT) is essential for sports analytics, enabling performance evaluation and tactical insights. However, tracking in sports is challenging due to fast movements, occlusions, and camera shifts. Traditional tracking-by-detection methods require extensive tuning, while segmentation-based approaches struggle with track processing. We propose McByte, a tracking-by-detection framework that integrates temporally propagated segmentation mask as an association cue to improve robustness without per-video tuning. Unlike many existing methods, McByte does not require training, relying solely on pre-trained models and object detectors commonly used in the community. Evaluated on SportsMOT, DanceTrack, SoccerNet-tracking 2022 and MOT17, McByte demonstrates strong performance across sports and general pedestrian tracking. Our results highlight the benefits of mask propagation for a more adaptable and generalizable MOT approach. Code will be made available at https://github.com/tstanczyk95/McByte.