BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-View
This work addresses the problem of efficient and accurate 3D object detection for autonomous driving systems, offering incremental improvements through optimized data augmentation and NMS strategies.
The paper tackles 3D object detection for autonomous driving by introducing the BEVDet paradigm, which operates in Bird-Eye-View and achieves high performance with efficient trade-offs, such as BEVDet-Tiny scoring 31.2% mAP and 39.2% NDS while running 9.2 times faster than FCOS3D, and BEVDet-Base scoring 39.3% mAP and 47.2% NDS, significantly exceeding prior results.
Autonomous driving perceives its surroundings for decision making, which is one of the most complex scenarios in visual perception. The success of paradigm innovation in solving the 2D object detection task inspires us to seek an elegant, feasible, and scalable paradigm for fundamentally pushing the performance boundary in this area. To this end, we contribute the BEVDet paradigm in this paper. BEVDet performs 3D object detection in Bird-Eye-View (BEV), where most target values are defined and route planning can be handily performed. We merely reuse existing modules to build its framework but substantially develop its performance by constructing an exclusive data augmentation strategy and upgrading the Non-Maximum Suppression strategy. In the experiment, BEVDet offers an excellent trade-off between accuracy and time-efficiency. As a fast version, BEVDet-Tiny scores 31.2% mAP and 39.2% NDS on the nuScenes val set. It is comparable with FCOS3D, but requires just 11% computational budget of 215.3 GFLOPs and runs 9.2 times faster at 15.6 FPS. Another high-precision version dubbed BEVDet-Base scores 39.3% mAP and 47.2% NDS, significantly exceeding all published results. With a comparable inference speed, it surpasses FCOS3D by a large margin of +9.8% mAP and +10.0% NDS. The source code is publicly available for further research at https://github.com/HuangJunJie2017/BEVDet .