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

arXiv cs.RO / 4/2/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

ProOOD is a lightweight, plug-and-play method for 3D semantic occupancy prediction that improves robustness to out-of-distribution (OOD) inputs and long-tailed class bias.
The approach uses prototype-guided semantic imputation to fill occluded regions with class-consistent features, helping reduce incorrect anomaly-to-rare-class assignments.
It introduces prototype-guided tail mining to strengthen rare-class representations and mitigate “OOD absorption” into tail classes.
ProOOD also proposes EchoOOD, which combines local logit coherence with local/global prototype matching to generate reliable voxel-level OOD scores.
Experiments on five datasets show state-of-the-art results, including gains on SemanticKITTI (+3.57% mIoU overall, +24.80% tail-class mIoU) and VAA-KITTI (+19.34 AuPRCr), with publicly available code.

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.