English

RePL: Pseudo-label Refinement for Semi-supervised LiDAR Semantic Segmentation

Computer Vision and Pattern Recognition 2026-04-09 v1

Abstract

Semi-supervised learning for LiDAR semantic segmentation often suffers from error propagation and confirmation bias caused by noisy pseudo-labels. To tackle this chronic issue, we introduce RePL, a novel framework that enhances pseudo-label quality by identifying and correcting potential errors in pseudo-labels through masked reconstruction, along with a dedicated training strategy. We also provide a theoretical analysis demonstrating the condition under which the pseudo-label refinement is beneficial, and empirically confirm that the condition is mild and clearly met by RePL. Extensive evaluations on the nuScenes-lidarseg and SemanticKITTI datasets show that RePL improves pseudo-label quality a lot and, as a result, achieves the state of the art in LiDAR semantic segmentation.

Keywords

Cite

@article{arxiv.2604.06825,
  title  = {RePL: Pseudo-label Refinement for Semi-supervised LiDAR Semantic Segmentation},
  author = {Donghyeon Kwon and Taegyu Park and Suha Kwak},
  journal= {arXiv preprint arXiv:2604.06825},
  year   = {2026}
}
R2 v1 2026-07-01T11:58:52.670Z