English

Gau-Occ: Geometry-Completed Gaussians for Multi-Modal 3D Occupancy Prediction

Computer Vision and Pattern Recognition 2026-03-25 v1

Abstract

3D semantic occupancy prediction is crucial for autonomous driving. While multi-modal fusion improves accuracy over vision-only methods, it typically relies on computationally expensive dense voxel or BEV tensors. We present Gau-Occ, a multi-modal framework that bypasses dense volumetric processing by modeling the scene as a compact collection of semantic 3D Gaussians. To ensure geometric completeness, we propose a LiDAR Completion Diffuser (LCD) that recovers missing structures from sparse LiDAR to initialize robust Gaussian anchors. Furthermore, we introduce Gaussian Anchor Fusion (GAF), which efficiently integrates multi-view image semantics via geometry-aligned 2D sampling and cross-modal alignment. By refining these compact Gaussian descriptors, Gau-Occ captures both spatial consistency and semantic discriminability. Extensive experiments across challenging benchmarks demonstrate that Gau-Occ achieves state-of-the-art performance with significant computational efficiency.

Keywords

Cite

@article{arxiv.2603.22852,
  title  = {Gau-Occ: Geometry-Completed Gaussians for Multi-Modal 3D Occupancy Prediction},
  author = {Chengxin Lv and Yihui Li and Hongyu Yang and YunHong Wang},
  journal= {arXiv preprint arXiv:2603.22852},
  year   = {2026}
}
R2 v1 2026-07-01T11:34:53.667Z