OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environments
This addresses the challenge of vision-based autonomous driving systems lacking LiDAR data, offering a novel approach for 3D environment reconstruction.
The paper tackles the problem of 3D occupancy prediction without LiDAR supervision by proposing OccNeRF, which uses neural rendering and multi-camera depth maps for training, achieving effective results on nuScenes and SemanticKITTI datasets.
Occupancy prediction reconstructs 3D structures of surrounding environments. It provides detailed information for autonomous driving planning and navigation. However, most existing methods heavily rely on the LiDAR point clouds to generate occupancy ground truth, which is not available in the vision-based system. In this paper, we propose an OccNeRF method for training occupancy networks without 3D supervision. Different from previous works which consider a bounded scene, we parameterize the reconstructed occupancy fields and reorganize the sampling strategy to align with the cameras' infinite perceptive range. The neural rendering is adopted to convert occupancy fields to multi-camera depth maps, supervised by multi-frame photometric consistency. Moreover, for semantic occupancy prediction, we design several strategies to polish the prompts and filter the outputs of a pretrained open-vocabulary 2D segmentation model. Extensive experiments for both self-supervised depth estimation and 3D occupancy prediction tasks on nuScenes and SemanticKITTI datasets demonstrate the effectiveness of our method.