BEVDepth: Acquisition of Reliable Depth for Multi-View 3D Object Detection

🤖 Plain-English Summary

In this research, we propose a new 3D object detector with a trustworthy depth estimation, dubbed BEVDepth, for camera-based Bird's-Eye-View~(BEV) 3D object detection. Aided by customized Efficient Voxel Pooling and multi-frame mechanism, BEVDepth achieves the new advanced 60.9% NDS on the challenging nuScenes test set while maintaining high efficiency.

🔑 Key Findings

• Our work is based on a key observation -- depth estimation in recent approaches is surprisingly inadequate given the fact that depth is essential to camera 3D detection.
• Our BEVDepth resolves this by leveraging explicit depth supervision.
• A camera-awareness depth estimation module is also introduced to facilitate the depth predicting capability.

🔗 Related Resources