Collaborative Tracking Learning for Frame-Rate-Insensitive Multi-Object Tracking

TL;DR

This paper introduces ColTrack, a collaborative learning approach for multi-object tracking that maintains high performance at low frame rates, reducing computational costs and improving efficiency on edge devices.

Contribution

The paper proposes a novel end-to-end query-based method with historical queries, an information refinement module, and a tracking object consistency loss for frame-rate-insensitive MOT.

Findings

Outperforms state-of-the-art on Dancetrack and BDD100K datasets.

Achieves higher accuracy than existing methods on MOT17.

Maintains high performance at reduced frame rates, enabling faster processing.

Abstract

Multi-object tracking (MOT) at low frame rates can reduce computational, storage and power overhead to better meet the constraints of edge devices. Many existing MOT methods suffer from significant performance degradation in low-frame-rate videos due to significant location and appearance changes between adjacent frames. To this end, we propose to explore collaborative tracking learning (ColTrack) for frame-rate-insensitive MOT in a query-based end-to-end manner. Multiple historical queries of the same target jointly track it with richer temporal descriptions. Meanwhile, we insert an information refinement module between every two temporal blocking decoders to better fuse temporal clues and refine features. Moreover, a tracking object consistency loss is proposed to guide the interaction between historical queries. Extensive experimental results demonstrate that in high-frame-rate videos, ColTrack obtains higher performance than state-of-the-art methods on large-scale datasets Dancetrack and BDD100K, and outperforms the existing end-to-end methods on MOT17. More importantly, ColTrack has a significant advantage over state-of-the-art methods in low-frame-rate videos, which allows it to obtain faster processing speeds by reducing frame-rate requirements while maintaining higher performance. Code will be released at https://github.com/yolomax/ColTrack