RopeBEV: A Multi-Camera Roadside Perception Network in Bird's-Eye-View
It solves roadside perception for autonomous driving systems, representing a novel application rather than an incremental improvement.
The paper tackles the lack of multi-camera Bird's-Eye-View perception for roadside autonomous driving by introducing RopeBEV, which addresses challenges like diverse camera poses and sparse regions, achieving first place on the RoScenes dataset and validation on a private dataset with over 600 cameras.
Multi-camera perception methods in Bird's-Eye-View (BEV) have gained wide application in autonomous driving. However, due to the differences between roadside and vehicle-side scenarios, there currently lacks a multi-camera BEV solution in roadside. This paper systematically analyzes the key challenges in multi-camera BEV perception for roadside scenarios compared to vehicle-side. These challenges include the diversity in camera poses, the uncertainty in Camera numbers, the sparsity in perception regions, and the ambiguity in orientation angles. In response, we introduce RopeBEV, the first dense multi-camera BEV approach. RopeBEV introduces BEV augmentation to address the training balance issues caused by diverse camera poses. By incorporating CamMask and ROIMask (Region of Interest Mask), it supports variable camera numbers and sparse perception, respectively. Finally, camera rotation embedding is utilized to resolve orientation ambiguity. Our method ranks 1st on the real-world highway dataset RoScenes and demonstrates its practical value on a private urban dataset that covers more than 50 intersections and 600 cameras.