3DLabelProp: Geometric-Driven Domain Generalization for LiDAR Semantic Segmentation in Autonomous Driving

TL;DR

3DLabelProp introduces a geometry-based domain generalization method for LiDAR semantic segmentation in autonomous driving, leveraging LiDAR sensor structure to improve robustness across diverse datasets, outperforming existing approaches.

Contribution

The paper presents a novel geometry-driven approach, 3DLabelProp, that enhances domain generalization in LiDAR semantic segmentation by exploiting LiDAR sensor structure, unlike traditional learning-based methods.

Findings

Outperforms existing domain generalization methods on seven datasets.

Demonstrates state-of-the-art performance in LiDAR semantic segmentation.

Validates the effectiveness of geometry-based approach over learning-based methods.

Abstract

Domain generalization aims to find ways for deep learning models to maintain their performance despite significant domain shifts between training and inference datasets. This is particularly important for models that need to be robust or are costly to train. LiDAR perception in autonomous driving is impacted by both of these concerns, leading to the emergence of various approaches. This work addresses the challenge by proposing a geometry-based approach, leveraging the sequential structure of LiDAR sensors, which sets it apart from the learning-based methods commonly found in the literature. The proposed method, called 3DLabelProp, is applied on the task of LiDAR Semantic Segmentation (LSS). Through extensive experimentation on seven datasets, it is demonstrated to be a state-of-the-art approach, outperforming both naive and other domain generalization methods.