HypLiLoc: Towards Effective LiDAR Pose Regression with Hyperbolic Fusion
Authors:
- 1 June 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5176-5185
Abstract
LiDAR relocalization plays a crucial role in many fields, including robotics, autonomous driving, and computer vision. LiDAR-based retrieval from a database typically incurs high computation storage costs and can lead to globally inaccurate pose estimations if the database is too sparse. On the other hand, pose regression methods take images or point clouds as inputs and directly regress global poses in an end-to-end manner. They do not perform database matching and are more computationally efficient than retrieval techniques. We propose HypLiLoc, a new model for LiDAR pose regression. We use two branched back-bones to extract 3D features and 2D projection features, respectively. We consider multi-modal feature fusion in both Euclidean and hyperbolic spaces to obtain more effective feature representations. Experimental results indicate that HypLiLoc achieves state-of-the-art performance in both outdoor and indoor datasets. We also conduct extensive ablation studies on the framework design, which demonstrate the effectiveness of multi-modal feature extraction and multi-space embedding. Our code is released at: https://github.com/sijieaaa/HypLiLocKeywords
All Related Versions
Funding Information
- A*Star (RIE2020)
- National Research Foundation, Singapore
This publication has 29 references indexed in Scilit:
- Ongoing Evolution of Visual SLAM from Geometry to Deep Learning: Challenges and OpportunitiesCognitive Computation, 2018
- PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition2018
- MapNet: An Allocentric Spatial Memory for Mapping Environments2018
- Image-Based Localization Using LSTMs for Structured Feature Correlation2017
- Topological map construction and scene recognition for vehicle localizationAutonomous Robots, 2017
- Deep Residual Learning for Image Recognition2016
- Low-drift and real-time lidar odometry and mappingAutonomous Robots, 2016
- PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization2015
- Low Distortion Delaunay Embedding of Trees in Hyperbolic Plane2012
- ImageNet: A large-scale hierarchical image database2009