Hongyu Yan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hongyu Yan has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Li Lu
Sichuan University, Chengdu, China
2 shared publications - Binqian Jiang
Department of Electronic and Computer Engineering, Hong Kong University of Science and Technology, Hong Kong, China
1 shared publication Chunchao Guo
1 shared publication- Di Xie
Hikvision Research Institute, Hangzhou, China
1 shared publication - Hang Du
Hikvision Research Institute Hangzhou China
1 shared publication Haolin Liu
1 shared publication
Latest Publications
- 5 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (5) , 4228-4235
MapEval: Towards Unified, Robust and Efficient SLAM Map Evaluation Framework
Evaluating massive-scale point cloud maps in Simultaneous Localization and Mapping (SLAM) still remains challenging due to three limitations: lack of unified standards, poor robustness to noise, and computational inefficiency. We propose MapEval, a novel framework for point cloud map assessment. Our key innovation is a voxelized Gaussian approximation method that enables efficient Wasserstein distance computation while maintaining physical meaning. This leads to two complementary metrics: Voxelized Average Wasserstein Distance (AWD) for global geometry and Spatial Consistency Score (SCS) for local consistency. Extensive experiments demonstrate that MapEval achieves $100$- $500$ times speedup while maintaining evaluation performance compared to traditional metrics like Chamfer Distance (CD) and Mean Map Entropy (MME). Our framework shows robust performance across both simulated and real-world datasets with million-scale point clouds.
- 5 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (5) , 4228-4235
MapEval: Towards Unified, Robust and Efficient SLAM Map Evaluation Framework
Evaluating massive-scale point cloud maps in Simultaneous Localization and Mapping (SLAM) still remains challenging due to three limitations: lack of unified standards, poor robustness to noise, and computational inefficiency. We propose MapEval, a novel framework for point cloud map assessment. Our key innovation is a voxelized Gaussian approximation method that enables efficient Wasserstein distance computation while maintaining physical meaning. This leads to two complementary metrics: Voxelized Average Wasserstein Distance (AWD) for global geometry and Spatial Consistency Score (SCS) for local consistency. Extensive experiments demonstrate that MapEval achieves $100$- $500$ times speedup while maintaining evaluation performance compared to traditional metrics like Chamfer Distance (CD) and Mean Map Entropy (MME). Our framework shows robust performance across both simulated and real-world datasets with million-scale point clouds.
- 11 November 2022
- Published by Springer Nature
- p. 676-693
FBNet: Feedback Network for Point Cloud Completion
- 11 November 2022
- Published by Springer Nature
- p. 676-693
FBNet: Feedback Network for Point Cloud Completion
- 7 September 2022
- Published by Springer Nature
- p. 557-569
Low-Level Graph Convolution Network for Point Cloud Processing
- 7 September 2022
- Published by Springer Nature
- p. 557-569
Low-Level Graph Convolution Network for Point Cloud Processing