5 papers
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…
Equivariant Neural Networks for General Linear Symmetries on Lie Algebras
Chankyo Kim, Sicheng Zhao, Minghan Zhu +2
Many scientific and geometric problems exhibit general linear symmetries, yet most equivariant neural networks are built for compact groups or simple vector features, limiting thei…
Equi-RO: A 4D mmWave Radar Odometry via Equivariant Networks
Zeyu Han, Shuocheng Yang, Minghan Zhu +4
Autonomous vehicles and robots rely on accurate odometry estimation in GPS-denied environments. While LiDARs and cameras struggle under extreme weather, 4D mmWave radar emerges as…
LiDAR-EDIT: LiDAR Data Generation by Editing the Object Layouts in Real-World Scenes
Shing-Hei Ho, Bao Thach, Minghan Zhu
We present LiDAR-EDIT, a novel paradigm for generating synthetic LiDAR data for autonomous driving. Our framework edits real-world LiDAR scans by introducing new object layouts whi…
LatentBKI: Open-Dictionary Continuous Mapping in Visual-Language Latent Spaces with Quantifiable Uncertainty
Joey Wilson, Ruihan Xu, Yile Sun +4
This paper introduces a novel probabilistic mapping algorithm, LatentBKI, which enables open-vocabulary mapping with quantifiable uncertainty. Traditionally, semantic mapping algor…