The reviewed record of science sign in
Pith

arxiv: 2304.13147 · v2 · pith:VXWN323C · submitted 2023-04-25 · cs.CV

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

Reviewed by Pithpith:VXWN323Copen to challenge →

classification cs.CV
keywords self-supervisedlearningre-identificationautonomousdrivingfeaturesframemethods
0
0 comments X
read the original 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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.