Unsupervised Domain Adaptation for 3D LiDAR Semantic Segmentation Using Contrastive Learning and Multi-Model Pseudo Labeling
It addresses the problem of costly manual annotation for autonomous systems by enabling robust adaptation across sensor and location shifts, though it is incremental as it builds on existing UDA and pseudo-labeling techniques.
This paper tackles performance degradation in 3D LiDAR semantic segmentation due to domain shifts by proposing a two-stage unsupervised domain adaptation framework, achieving significant improvements in segmentation accuracy on datasets like SemanticKITTI to SemanticPOSS and SemanticSlamantic compared to baseline methods.
Addressing performance degradation in 3D LiDAR semantic segmentation due to domain shifts (e.g., sensor type, geographical location) is crucial for autonomous systems, yet manual annotation of target data is prohibitive. This study addresses the challenge using Unsupervised Domain Adaptation (UDA) and introduces a novel two-stage framework to tackle it. Initially, unsupervised contrastive learning at the segment level is used to pre-train a backbone network, enabling it to learn robust, domain-invariant features without labels. Subsequently, a multi-model pseudo-labeling strategy is introduced, utilizing an ensemble of diverse state-of-the-art architectures (including projection, voxel, hybrid, and cylinder-based methods). Predictions from these models are aggregated via hard voting to generate high-quality, refined pseudo-labels for the unlabeled target domain, mitigating single-model biases. The contrastively pre-trained network is then fine-tuned using these robust pseudo-labels. Experiments adapting from SemanticKITTI to unlabeled target datasets (SemanticPOSS, SemanticSlamantic) demonstrate significant improvements in segmentation accuracy compared to direct transfer and single-model UDA approaches. These results highlight the effectiveness of combining contrastive pre-training with refined ensemble pseudo-labeling for bridging complex domain gaps without requiring target domain annotations.