FB-OCC: 3D Occupancy Prediction based on Forward-Backward View Transformation
This work addresses 3D occupancy prediction for autonomous driving systems, representing an incremental improvement over existing methods.
The paper tackles 3D occupancy prediction for autonomous driving by proposing FB-OCC, which builds on FB-BEV with tailored designs like joint depth-semantic pre-training and joint voxel-BEV representation, achieving a state-of-the-art mIoU score of 54.19% on the nuScenes dataset and winning the CVPR 2023 challenge.
This technical report summarizes the winning solution for the 3D Occupancy Prediction Challenge, which is held in conjunction with the CVPR 2023 Workshop on End-to-End Autonomous Driving and CVPR 23 Workshop on Vision-Centric Autonomous Driving Workshop. Our proposed solution FB-OCC builds upon FB-BEV, a cutting-edge camera-based bird's-eye view perception design using forward-backward projection. On top of FB-BEV, we further study novel designs and optimization tailored to the 3D occupancy prediction task, including joint depth-semantic pre-training, joint voxel-BEV representation, model scaling up, and effective post-processing strategies. These designs and optimization result in a state-of-the-art mIoU score of 54.19% on the nuScenes dataset, ranking the 1st place in the challenge track. Code and models will be released at: https://github.com/NVlabs/FB-BEV.