Towards a Principled Integration of Multi-Camera Re-Identification and Tracking through Optimal Bayes Filters

TL;DR

This paper proposes a theoretically principled method to integrate multi-camera person re-identification with tracking using optimal Bayes filters, aiming for a more direct and end-to-end approach from raw images.

Contribution

It introduces a novel, Bayes filter-based framework that seamlessly combines ReID and tracking, reducing reliance on data association and enabling direct processing from raw images.

Findings

Initial results show promise but are still sub-par.

The approach simplifies the tracking pipeline.

Opens new research directions for end-to-end multi-camera tracking.

Abstract

With the rise of end-to-end learning through deep learning, person detectors and re-identification (ReID) models have recently become very strong. Multi-camera multi-target (MCMT) tracking has not fully gone through this transformation yet. We intend to take another step in this direction by presenting a theoretically principled way of integrating ReID with tracking formulated as an optimal Bayes filter. This conveniently side-steps the need for data-association and opens up a direct path from full images to the core of the tracker. While the results are still sub-par, we believe that this new, tight integration opens many interesting research opportunities and leads the way towards full end-to-end tracking from raw pixels.