JDT3D: Addressing the Gaps in LiDAR-Based Tracking-by-Attention
This work addresses a specific bottleneck in 3D tracking for autonomous driving, offering incremental improvements to TBA methods.
The paper tackled the performance gap between tracking-by-attention (TBA) and tracking-by-detection (TBD) methods in 3D LiDAR-based tracking for autonomous driving, proposing JDT3D with track sampling augmentation and confidence-based query propagation, which achieved 0.574 AMOTA on nuScenes, outperforming existing TBA methods by over 6%.
Tracking-by-detection (TBD) methods achieve state-of-the-art performance on 3D tracking benchmarks for autonomous driving. On the other hand, tracking-by-attention (TBA) methods have the potential to outperform TBD methods, particularly for long occlusions and challenging detection settings. This work investigates why TBA methods continue to lag in performance behind TBD methods using a LiDAR-based joint detector and tracker called JDT3D. Based on this analysis, we propose two generalizable methods to bridge the gap between TBD and TBA methods: track sampling augmentation and confidence-based query propagation. JDT3D is trained and evaluated on the nuScenes dataset, achieving 0.574 on the AMOTA metric on the nuScenes test set, outperforming all existing LiDAR-based TBA approaches by over 6%. Based on our results, we further discuss some potential challenges with the existing TBA model formulation to explain the continued gap in performance with TBD methods. The implementation of JDT3D can be found at the following link: https://github.com/TRAILab/JDT3D.