Unsupervised Domain Adaptation for LiDAR Panoptic Segmentation

TL;DR

This paper introduces AdaptLPS, a novel unsupervised domain adaptation method for LiDAR panoptic segmentation, which improves model performance across different sensor setups without additional data labeling.

Contribution

AdaptLPS combines data-based and model-based domain adaptation strategies specifically tailored for LiDAR panoptic segmentation, addressing sensor and environmental variations.

Findings

Outperforms existing UDA methods by up to 6.41 percentage points in PQ score

Effectively handles variations in LiDAR sensor configurations and environmental conditions

Demonstrates robustness across multiple real-world autonomous driving datasets

Abstract

Scene understanding is a pivotal task for autonomous vehicles to safely navigate in the environment. Recent advances in deep learning enable accurate semantic reconstruction of the surroundings from LiDAR data. However, these models encounter a large domain gap while deploying them on vehicles equipped with different LiDAR setups which drastically decreases their performance. Fine-tuning the model for every new setup is infeasible due to the expensive and cumbersome process of recording and manually labeling new data. Unsupervised Domain Adaptation (UDA) techniques are thus essential to fill this domain gap and retain the performance of models on new sensor setups without the need for additional data labeling. In this paper, we propose AdaptLPS, a novel UDA approach for LiDAR panoptic segmentation that leverages task-specific knowledge and accounts for variation in the number of scan lines, mounting position, intensity distribution, and environmental conditions. We tackle the UDA task by employing two complementary domain adaptation strategies, data-based and model-based. While data-based adaptations reduce the domain gap by processing the raw LiDAR scans to resemble the scans in the target domain, model-based techniques guide the network in extracting features that are representative for both domains. Extensive evaluations on three pairs of real-world autonomous driving datasets demonstrate that AdaptLPS outperforms existing UDA approaches by up to 6.41 pp in terms of the PQ score.