5 papers
CVKD-UDA: Cross-View Knowledge Distillation for 3D Unsupervised Domain Adaptive Segmentation
Zhimin Yuan, Ming Cheng, Shangshu Yu +4
3D unsupervised domain adaptive (UDA) segmentation mitigates the high cost of manual annotations of the new domain data. Self-training has emerged as the dominant approach in this…
LEADER: Learning Reliable Local-to-Global Correspondences for LiDAR Relocalization
Jianshi Wu, Minghang Zhu, Dunqiang Liu +5
LiDAR relocalization has attracted increasing attention as it can deliver accurate 6-DoF pose estimation in complex 3D environments. Recent learning-based regression methods offer…
From Single Scan to Sequential Consistency: A New Paradigm for LIDAR Relocalization
Minghang Zhu, Zhijing Wang, Yuxin Guo +3
LiDAR relocalization aims to estimate the global 6-DoF pose of a sensor in the environment. However, existing regression-based approaches are prone to dynamic or ambiguous scenario…
V2VLoc: Robust GNSS-Free Collaborative Perception via LiDAR Localization
Wenkai Lin, Qiming Xia, Wen Li +2
Multi-agents rely on accurate poses to share and align observations, enabling a collaborative perception of the environment. However, traditional GNSS-based localization often fail…
LightLoc: Learning Outdoor LiDAR Localization at Light Speed
Wen Li, Chen Liu, Shangshu Yu +5
Scene coordinate regression achieves impressive results in outdoor LiDAR localization but requires days of training. Since training needs to be repeated for each new scene, long tr…