Uncertainty-aware LiDAR Panoptic Segmentation

TL;DR

This paper introduces EvLPSNet, a novel LiDAR panoptic segmentation network that efficiently predicts semantic, instance, and uncertainty information in a sampling-free manner, improving scene understanding for autonomous systems.

Contribution

EvLPSNet is the first to perform uncertainty-aware LiDAR panoptic segmentation efficiently without sampling, enhancing reliability in autonomous scene understanding.

Findings

Achieves state-of-the-art uncertainty-aware segmentation quality.

Provides well-calibrated uncertainty estimates.

Demonstrates improved performance over baselines.

Abstract

Modern autonomous systems often rely on LiDAR scanners, in particular for autonomous driving scenarios. In this context, reliable scene understanding is indispensable. Current learning-based methods typically try to achieve maximum performance for this task, while neglecting a proper estimation of the associated uncertainties. In this work, we introduce a novel approach for solving the task of uncertainty-aware panoptic segmentation using LiDAR point clouds. Our proposed EvLPSNet network is the first to solve this task efficiently in a sampling-free manner. It aims to predict per-point semantic and instance segmentations, together with per-point uncertainty estimates. Moreover, it incorporates methods for improving the performance by employing the predicted uncertainties. We provide several strong baselines combining state-of-the-art panoptic segmentation networks with sampling-free uncertainty estimation techniques. Extensive evaluations show that we achieve the best performance on uncertainty-aware panoptic segmentation quality and calibration compared to these baselines. We make our code available at: https://github.com/kshitij3112/EvLPSNet