BEVTrack: A Simple and Strong Baseline for 3D Single Object Tracking in Bird's-Eye View

TL;DR

BEVTrack introduces a simple, motion-based 3D object tracking method in Bird's-Eye View that adapts likelihood functions for diverse targets, achieving state-of-the-art accuracy and real-time performance.

Contribution

The paper presents BEVTrack, a novel, straightforward approach that estimates object motion directly in BEV and learns adaptive likelihood functions, improving robustness and accuracy over complex existing methods.

Findings

Achieves state-of-the-art results on KITTI, NuScenes, and Waymo datasets.

Operates at 200 FPS, enabling real-time tracking.

Effectively handles diverse target attributes with adaptive likelihood learning.

Abstract

3D Single Object Tracking (SOT) is a fundamental task in computer vision and plays a critical role in applications like autonomous driving. However, existing algorithms often involve complex designs and multiple loss functions, making model training and deployment challenging. Furthermore, their reliance on fixed probability distribution assumptions (e.g., Laplacian or Gaussian) hinders their ability to adapt to diverse target characteristics such as varying sizes and motion patterns, ultimately affecting tracking precision and robustness. To address these issues, we propose BEVTrack, a simple yet effective motion-based tracking method. BEVTrack directly estimates object motion in Bird's-Eye View (BEV) using a single regression loss. To enhance accuracy for targets with diverse attributes, it learns adaptive likelihood functions tailored to individual targets, avoiding the limitations of fixed distribution assumptions in previous methods. This approach provides valuable priors for tracking and significantly boosts performance. Comprehensive experiments on KITTI, NuScenes, and Waymo Open Dataset demonstrate that BEVTrack achieves state-of-the-art results while operating at 200 FPS, enabling real-time applicability. The code will be released at https://github.com/xmm-prio/BEVTrack.