$\alpha$-OCC: Uncertainty-Aware Camera-based 3D Semantic Occupancy Prediction

TL;DR

This paper introduces $oldsymbol{ ext{ extalpha}- ext{OCC}}$, an uncertainty-aware camera-based 3D semantic occupancy prediction method that enhances geometry completion and uncertainty quantification for autonomous vehicle perception.

Contribution

It proposes Depth-UP for uncertainty propagation and HCP for hierarchical conformal prediction, significantly improving OCC accuracy and uncertainty handling in autonomous perception systems.

Findings

Depth-UP improves geometry completion by up to 11.58%

HCP reduces prediction set size by up to 92% while maintaining coverage

KL-based score enhances safety-critical class recall by 45%

Abstract

In the realm of autonomous vehicle perception, comprehending 3D scenes is paramount for tasks such as planning and mapping. Camera-based 3D Semantic Occupancy Prediction (OCC) aims to infer scene geometry and semantics from limited observations. While it has gained popularity due to affordability and rich visual cues, existing methods often neglect the inherent uncertainty in models. To address this, we propose an uncertainty-aware OCC method ($\alpha$-OCC). We first introduce Depth-UP, an uncertainty propagation framework that improves geometry completion by up to 11.58\% and semantic segmentation by up to 12.95\% across various OCC models. For uncertainty quantification (UQ), we propose the hierarchical conformal prediction (HCP) method, effectively handling the high-level class imbalance in OCC datasets. On the geometry level, the novel KL-based score function significantly improves the occupied recall (45\%) of safety-critical classes with minimal performance overhead (3.4\% reduction). On UQ, our HCP achieves smaller prediction set sizes while maintaining the defined coverage guarantee. Compared with baselines, it reduces up to 92\% set size, with 18\% further reduction when integrated with Depth-UP. Our contributions advance OCC accuracy and robustness, marking a noteworthy step forward in autonomous perception systems.