Occlusion-Aware Detection and Re-ID Calibrated Network for Multi-Object Tracking

TL;DR

This paper introduces ORCTrack, a novel multi-object tracking network that effectively handles occlusions by integrating an occlusion-aware attention module and a Re-ID calibration method, achieving state-of-the-art results.

Contribution

The paper proposes an occlusion-aware detection and Re-ID calibration network with a new attention module and Re-ID matching block, addressing occlusion challenges in MOT.

Findings

Achieves new state-of-the-art performance on VisDrone2021-MOT and KITTI benchmarks.

Demonstrates improved detection and Re-ID accuracy under occlusion conditions.

Maintains high run-time efficiency while enhancing tracking robustness.

Abstract

Multi-Object Tracking (MOT) is a crucial computer vision task that aims to predict the bounding boxes and identities of objects simultaneously. While state-of-the-art methods have made remarkable progress by jointly optimizing the multi-task problems of detection and Re-ID feature learning, yet, few approaches explore to tackle the occlusion issue, which is a long-standing challenge in the MOT field. Generally, occluded objects may hinder the detector from estimating the bounding boxes, resulting in fragmented trajectories. And the learned occluded Re-ID embeddings are less distinct since they contain interferer. To this end, we propose an occlusion-aware detection and Re-ID calibrated network for multi-object tracking, termed as ORCTrack. Specifically, we propose an Occlusion-Aware Attention (OAA) module in the detector that highlights the object features while suppressing the occluded background regions. OAA can serve as a modulator that enhances the detector for some potentially occluded objects. Furthermore, we design a Re-ID embedding matching block based on the optimal transport problem, which focuses on enhancing and calibrating the Re-ID representations through different adjacent frames complementarily. To validate the effectiveness of the proposed method, extensive experiments are conducted on two challenging VisDrone2021-MOT and KITTI benchmarks. Experimental evaluations demonstrate the superiority of our approach, which can achieve new state-of-the-art performance and enjoy high run-time efficiency.