Inferring the 3D structure of a scene from a single image is an ill-posed and challenging problem in the field of vision-centric autonomous driving. Existing methods usually employ neural radiance fields to produce voxelized 3D occupancy, lacking instance-level semantic reasoning and temporal photometric consistency. In this paper, we propose ViPOcc, which leverages the visual priors from vision foundation models (VFMs) for fine-grained 3D occupancy prediction. Unlike previous works that solely employ volume rendering for RGB and depth image reconstruction, we introduce a metric depth estimation branch, in which an inverse depth alignment module is proposed to bridge the domain gap in depth distribution between VFM predictions and the ground truth. The recovered metric depth is then utilized in temporal photometric alignment and spatial geometric alignment to ensure accurate and consistent 3D occupancy prediction. Additionally, we also propose a semantic-guided non-overlapping Gaussian mixture sampler for efficient, instance-aware ray sampling, which addresses the redundant and imbalanced sampling issue that still exists in previous state-of-the-arts. Extensive experiments demonstrate the superior performance of ViPOcc in both 3D occupancy prediction and depth estimation tasks on the KITTI-360 and KITTI Raw datasets. Our source code will be released upon publication.
An illustration of our proposed ViPOcc framework. Unlike previous approaches that rely solely on NeRF for 3D scene reconstruction, ViPOcc introduces an additional depth prediction branch and a novel SNOG sampler for temporal photometric alignment and spatial geometric alignment.
An illustration of our proposed SNOG sampler. Our innovative SNOG sampler prioritizes key instances and eliminates overlapping patches, optimizing ray sampling for efficient and accurate 3D occupancy prediction. By addressing redundancy and enhancing instance representation, the SNOG sampler sets a new standard for semantic-guided sampling in 3D vision.
@inproceedings{feng2025vipocc,
title={{ViPOcc: Leveraging Visual Priors from Vision Foundation Models for Single-View 3D Occupancy Prediction}},
author={Feng, Yi and others},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
year={2025},
}