English

Self-Supervised Multi-Object Tracking For Autonomous Driving From Consistency Across Timescales

Computer Vision and Pattern Recognition 2023-09-22 v2

Abstract

Self-supervised multi-object trackers have tremendous potential as they enable learning from raw domain-specific data. However, their re-identification accuracy still falls short compared to their supervised counterparts. We hypothesize that this drawback results from formulating self-supervised objectives that are limited to single frames or frame pairs. Such formulations do not capture sufficient visual appearance variations to facilitate learning consistent re-identification features for autonomous driving when the frame rate is low or object dynamics are high. In this work, we propose a training objective that enables self-supervised learning of re-identification features from multiple sequential frames by enforcing consistent association scores across short and long timescales. We perform extensive evaluations demonstrating that re-identification features trained from longer sequences significantly reduce ID switches on standard autonomous driving datasets compared to existing self-supervised learning methods, which are limited to training on frame pairs. Using our proposed SubCo loss function, we set the new state-of-the-art among self-supervised methods and even perform on par with fully supervised learning methods.

Keywords

Cite

@article{arxiv.2304.13147,
  title  = {Self-Supervised Multi-Object Tracking For Autonomous Driving From Consistency Across Timescales},
  author = {Christopher Lang and Alexander Braun and Lars Schillingmann and Abhinav Valada},
  journal= {arXiv preprint arXiv:2304.13147},
  year   = {2023}
}

Comments

8 pages, 3 figures, 5 tables

R2 v1 2026-06-28T10:17:48.289Z