ProOOD: Prototype-Guided Out-of-Distribution 3D Occupancy Prediction

TL;DR

ProOOD introduces a novel prototype-guided approach for 3D occupancy prediction that enhances in-distribution accuracy and out-of-distribution detection, especially for rare classes, in autonomous driving scenarios.

Contribution

It proposes a lightweight, plug-and-play method combining prototype-guided refinement and training-free OOD scoring for improved 3D occupancy prediction and OOD detection.

Findings

Achieves state-of-the-art results on five datasets.

Surpasses baselines by +3.57% mIoU on SemanticKITTI.

Improves AuPRCr by +19.34 points on VAA-KITTI.

Abstract

3D semantic occupancy prediction is central to autonomous driving, yet current methods are vulnerable to long-tailed class bias and out-of-distribution (OOD) inputs, often overconfidently assigning anomalies to rare classes. We present ProOOD, a lightweight, plug-and-play method that couples prototype-guided refinement with training-free OOD scoring. ProOOD comprises (i) prototype-guided semantic imputation that fills occluded regions with class-consistent features, (ii) prototype-guided tail mining that strengthens rare-class representations to curb OOD absorption, and (iii) EchoOOD, which fuses local logit coherence with local and global prototype matching to produce reliable voxel-level OOD scores. Extensive experiments on five datasets demonstrate that ProOOD achieves state-of-the-art performance on both in-distribution 3D occupancy prediction and OOD detection. On SemanticKITTI, it surpasses baselines by +3.57% mIoU overall and +24.80% tail-class mIoU; on VAA-KITTI, it improves AuPRCr by +19.34 points, with consistent gains across benchmarks. These improvements yield more calibrated occupancy estimates and more reliable OOD detection in safety-critical urban driving. The source code is publicly available at https://github.com/7uHeng/ProOOD.