Robust Single Object Tracking in LiDAR Point Clouds under Adverse Weather Conditions

TL;DR

This paper introduces a new adverse weather benchmark for LiDAR-based 3D single object tracking, evaluates existing methods' robustness under challenging conditions, and proposes a dual-branch framework, DRCT, that significantly improves tracking performance in adverse weather.

Contribution

It provides the first comprehensive adverse weather benchmark for LiDAR 3DSOT and develops a novel dual-branch tracking framework, DRCT, that enhances robustness in such conditions.

Findings

Existing trackers suffer significant performance drops in adverse weather.

Adverse weather impacts target distance and shape, degrading tracking accuracy.

DRCT outperforms baseline methods in adverse weather benchmarks.

Abstract

3D single object tracking (3DSOT) in LiDAR point clouds is a critical task for outdoor perception, enabling real-time perception of object location, orientation, and motion. Despite the impressive performance of current 3DSOT methods, evaluating them on clean datasets inadequately reflects their comprehensive performance, as the adverse weather conditions in real-world surroundings has not been considered. One of the main obstacles is the lack of adverse weather benchmarks for the evaluation of 3DSOT. To this end, this work proposes a challenging benchmark for LiDAR-based 3DSOT in adverse weather, which comprises two synthetic datasets (KITTI-A and nuScenes-A) and one real-world dataset (CADC-SOT) spanning three weather types: rain, fog, and snow. Based on this benchmark, five representative 3D trackers from different tracking frameworks conducted robustness evaluation, resulting in significant performance degradations. This prompts the question: What are the factors that cause current advanced methods to fail on such adverse weather samples? Consequently, we explore the impacts of adverse weather and answer the above question from three perspectives: 1) target distance; 2) template shape corruption; and 3) target shape corruption. Finally, based on domain randomization and contrastive learning, we designed a dual-branch tracking framework for adverse weather, named DRCT, achieving excellent performance in benchmarks.