LiOn-XA: Unsupervised Domain Adaptation via LiDAR-Only Cross-Modal Adversarial Training

TL;DR

LiOn-XA introduces an unsupervised domain adaptation method for 3D LiDAR segmentation that leverages cross-modal learning between 3D point clouds and 2D range images without relying on RGB data, using adversarial training to improve domain transfer.

Contribution

The paper presents a novel LiDAR-only cross-modal adversarial training approach for unsupervised domain adaptation in 3D segmentation, effective without RGB images.

Findings

Achieves state-of-the-art performance on real-to-real LiDAR domain adaptation tasks.

Effectively learns from two LiDAR data representations to bridge domain gaps.

Demonstrates robustness across multiple adaptation scenarios.

Abstract

In this paper, we propose LiOn-XA, an unsupervised domain adaptation (UDA) approach that combines LiDAR-Only Cross-Modal (X) learning with Adversarial training for 3D LiDAR point cloud semantic segmentation to bridge the domain gap arising from environmental and sensor setup changes. Unlike existing works that exploit multiple data modalities like point clouds and RGB image data, we address UDA in scenarios where RGB images might not be available and show that two distinct LiDAR data representations can learn from each other for UDA. More specifically, we leverage 3D voxelized point clouds to preserve important geometric structure in combination with 2D projection-based range images that provide information such as object orientations or surfaces. To further align the feature space between both domains, we apply adversarial training using both features and predictions of both 2D and 3D neural networks. Our experiments on 3 real-to-real adaptation scenarios demonstrate the effectiveness of our approach, achieving new state-of-the-art performance when compared to previous uni- and multi-model UDA methods. Our source code is publicly available at https://github.com/JensLe97/lion-xa.