PieTrack: An MOT solution based on synthetic data training and self-supervised domain adaptation

TL;DR

PieTrack is a multi-object tracking solution that leverages synthetic data and self-supervised domain adaptation to effectively track humans without pre-trained models, achieving competitive results on MOT17.

Contribution

The paper introduces a novel self-supervised domain adaptation method for MOT that does not require human labels, improving synthetic-to-real transfer.

Findings

Achieved a HOTA score of 58.7 on MOT17

Ranked third in the BMTT challenge

Effective domain adaptation without extra human labels

Abstract

In order to cope with the increasing demand for labeling data and privacy issues with human detection, synthetic data has been used as a substitute and showing promising results in human detection and tracking tasks. We participate in the 7th Workshop on Benchmarking Multi-Target Tracking (BMTT), themed on "How Far Can Synthetic Data Take us"? Our solution, PieTrack, is developed based on synthetic data without using any pre-trained weights. We propose a self-supervised domain adaptation method that enables mitigating the domain shift issue between the synthetic (e.g., MOTSynth) and real data (e.g., MOT17) without involving extra human labels. By leveraging the proposed multi-scale ensemble inference, we achieved a final HOTA score of 58.7 on the MOT17 testing set, ranked third place in the challenge.