BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detection
This work addresses a domain-specific problem for autonomous driving systems by enhancing roadside camera perception, representing an incremental advancement over existing methods.
The paper tackles the problem of inferior performance of vision-centric bird's eye view detection methods on roadside cameras by proposing BEVHeight, which regresses height to the ground instead of depth, achieving significant improvements on popular 3D detection benchmarks.
While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird's eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight, to address this issue. In essence, instead of predicting the pixel-wise depth, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. The code is available at {\url{https://github.com/ADLab-AutoDrive/BEVHeight}}.