MotionTrack: Learning Motion Predictor for Multiple Object Tracking

TL;DR

MotionTrack introduces a learnable, motion-based predictor utilizing self-attention and dynamic MLPs to improve multi-object tracking accuracy in complex motion scenarios, surpassing existing methods.

Contribution

The paper presents a novel online tracking method with a learnable motion predictor that effectively models temporal dynamics using multi-granularity motion features.

Findings

Achieves state-of-the-art results on Dancetrack and SportsMOT datasets.

Effectively models complex object motion with a self-attention mechanism.

Outperforms traditional linear motion models in challenging scenarios.

Abstract

Significant progress has been achieved in multi-object tracking (MOT) through the evolution of detection and re-identification (ReID) techniques. Despite these advancements, accurately tracking objects in scenarios with homogeneous appearance and heterogeneous motion remains a challenge. This challenge arises from two main factors: the insufficient discriminability of ReID features and the predominant utilization of linear motion models in MOT. In this context, we introduce a novel motion-based tracker, MotionTrack, centered around a learnable motion predictor that relies solely on object trajectory information. This predictor comprehensively integrates two levels of granularity in motion features to enhance the modeling of temporal dynamics and facilitate precise future motion prediction for individual objects. Specifically, the proposed approach adopts a self-attention mechanism to capture token-level information and a Dynamic MLP layer to model channel-level features. MotionTrack is a simple, online tracking approach. Our experimental results demonstrate that MotionTrack yields state-of-the-art performance on datasets such as Dancetrack and SportsMOT, characterized by highly complex object motion.