FB-OCC：基于前向-后向视图变换的3D占用预测
FB-OCC: 3D Occupancy Prediction based on Forward-Backward View Transformation
AI总结：
本文提出FB-OCC，在FB-BEV基础上通过联合深度-语义预训练、体素-BEV表示、模型扩展和后处理优化，在nuScenes上以54.19% mIoU夺得3D占用预测挑战赛第一名。
摘要：
本技术报告总结了3D占用预测挑战赛的获奖解决方案，该挑战赛与CVPR 2023端到端自动驾驶研讨会和CVPR 23视觉中心自动驾驶研讨会联合举办。我们提出的解决方案FB-OCC建立在FB-BEV之上，FB-BEV是一种使用前向-后向投影的基于摄像头的先进鸟瞰图感知设计。在FB-BEV的基础上，我们进一步研究了针对3D占用预测任务量身定制的新颖设计和优化，包括联合深度-语义预训练、联合体素-BEV表示、模型规模扩展以及有效的后处理策略。这些设计和优化在nuScenes数据集上取得了54.19%的最先进mIoU分数，在该挑战赛赛道中排名第一。代码和模型将在以下地址发布：https://github.com/NVlabs/FB-BEV。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
