RePL refines pseudo-labels for semi-supervised LiDAR semantic segmentation via masked reconstruction, achieving state-of-the-art results on nuScenes-lidarseg and SemanticKITTI.
Panoptic nuscenes: A large-scale benchmark for lidar panoptic segmentation and tracking.arXiv preprint arXiv:2109.03805
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Satellite imagery is integrated via cross-view attention into feed-forward 3D reconstruction to resolve global scale ambiguity and produce metric outputs.
A query-driven digital twin for autonomous driving reduces planning position error by 24% and communication overhead by 40% using optimization and progressive querying.
citing papers explorer
-
RePL: Pseudo-label Refinement for Semi-supervised LiDAR Semantic Segmentation
RePL refines pseudo-labels for semi-supervised LiDAR semantic segmentation via masked reconstruction, achieving state-of-the-art results on nuScenes-lidarseg and SemanticKITTI.
-
Empowering Feed-Forward Reconstruction Models with Metric Scale via Satellite Images
Satellite imagery is integrated via cross-view attention into feed-forward 3D reconstruction to resolve global scale ambiguity and produce metric outputs.