Boosting Online 3D Multi-Object Tracking through Camera-Radar Cross Check

TL;DR

This paper introduces CRAFTBooster, a novel sensor fusion method combining camera and radar data to significantly improve 3D multi-object tracking accuracy in autonomous driving.

Contribution

It presents a pioneering radar-camera fusion approach at the tracking stage, enhancing 3D MOT performance beyond existing single-modality methods.

Findings

Achieved 5-6% IDF1 performance gain on K-Radaar dataset

Demonstrated superior detection and tracking accuracy

Validated effectiveness of sensor fusion in autonomous driving

Abstract

In the domain of autonomous driving, the integration of multi-modal perception techniques based on data from diverse sensors has demonstrated substantial progress. Effectively surpassing the capabilities of state-of-the-art single-modality detectors through sensor fusion remains an active challenge. This work leverages the respective advantages of cameras in perspective view and radars in Bird's Eye View (BEV) to greatly enhance overall detection and tracking performance. Our approach, Camera-Radar Associated Fusion Tracking Booster (CRAFTBooster), represents a pioneering effort to enhance radar-camera fusion in the tracking stage, contributing to improved 3D MOT accuracy. The superior experimental results on the K-Radaar dataset, which exhibit 5-6% on IDF1 tracking performance gain, validate the potential of effective sensor fusion in advancing autonomous driving.