Search papers, labs, and topics across Lattice.
This paper introduces a teacher-student semi-supervised learning framework that addresses the challenge of limited labeled data in online HD map construction by generating high-quality pseudo-labels through confidence-aware map refinement. The method employs Beta-distribution-based confidence maps to evaluate the reliability of predicted map elements, utilizing a spatial clipping technique to selectively retain high-confidence regions while discarding unreliable ones. Experimental results on the nuScenes dataset indicate that this approach enhances prediction accuracy and achieves a +6.1 mAP improvement under low-label conditions compared to traditional labeled data training.
Confidence-aware pseudo-labeling boosts performance in HD map construction, achieving a +6.1 mAP gain even with scarce labeled data.
A critical challenge in deploying online HD map construction systems to real-world scenarios is the scarcity of labeled training data, which limits model generalization in diverse environments. To address this limitation, we propose a teacher-student semi-supervised learning (SSL) framework that generates high-quality pseudo-labels from unlabeled data through confidence-aware map refinement. Our approach first trains a teacher model on limited labeled data, then leverages Beta-distribution-based confidence maps to assess the reliability of predicted map elements across temporal observations. Unlike conventional filtering methods that discard entire elements, we introduce a spatial clipping technique that selectively preserves high-confidence regions while removing unreliable segments. The refined map elements serve as map priors that improve the teacher model's prediction accuracy on unlabeled data in a second pass. These enhanced predictions become pseudo-labels for training a student model from scratch, followed by fine-tuning on the original labeled data. Experimental results on the nuScenes dataset demonstrate that our teacher-student framework with refined pseudo-labels improves performance by +6.1 mAP under a low-label regime compared to training on labeled data alone, offering a practical solution to the labeled data scarcity problem in online HD map construction.
