Online Multi-Object Tracking with Unsupervised Re-Identification Learning and Occlusion Estimation

TL;DR

This paper introduces unsupervised re-identification and occlusion estimation modules for online multi-object tracking, effectively handling occlusions and missed detections without requiring identity annotations, thereby improving tracking accuracy.

Contribution

It presents novel unsupervised re-identification and occlusion estimation modules that enhance online multi-object tracking, especially for highly occluded objects, without needing identity labels.

Findings

Unsupervised re-identification performs comparably to supervised methods.

Occlusion estimation improves tracking of missed objects.

Tracking performance is significantly enhanced with the combined modules.

Abstract

Occlusion between different objects is a typical challenge in Multi-Object Tracking (MOT), which often leads to inferior tracking results due to the missing detected objects. The common practice in multi-object tracking is re-identifying the missed objects after their reappearance. Though tracking performance can be boosted by the re-identification, the annotation of identity is required to train the model. In addition, such practice of re-identification still can not track those highly occluded objects when they are missed by the detector. In this paper, we focus on online multi-object tracking and design two novel modules, the unsupervised re-identification learning module and the occlusion estimation module, to handle these problems. Specifically, the proposed unsupervised re-identification learning module does not require any (pseudo) identity information nor suffer from the scalability issue. The proposed occlusion estimation module tries to predict the locations where occlusions happen, which are used to estimate the positions of missed objects by the detector. Our study shows that, when applied to state-of-the-art MOT methods, the proposed unsupervised re-identification learning is comparable to supervised re-identification learning, and the tracking performance is further improved by the proposed occlusion estimation module.