Hongyu Sun
Dr. Hongyu Sun
Email GitHub ORCID Scholar LinkedIn

About Me

I am a researcher and faculty member at the College of Computer Science, Inner Mongolia University, China, where I lead the Spatial AI Lab. My research focuses on 3D Computer Vision, 3D Perception, and Large Multi-modal Models, with particular interests in point cloud processing, open-world learning, and practical AI systems.

I received my Ph.D. in Computer Science (2025) from Renmin University of China, supervised by Prof. Deying Li and Prof. Yongcai Wang. My academic journey includes a Master's degree from Renmin University (2017-2020) and a Bachelor's degree from Inner Mongolia University (2013-2017).

Research Interests

Spatial AI 3D Computer Vision 3D Perception Point Cloud Processing Large Multi-modal Models Open-world Learning Practical Applications

News

January 2026
Paper Accepted: "Boot-and-Feedback Framework for Generalist-Expert Model Collaboration in Breast Ultrasound Diagnosis" accepted by IEEE ICASSP 2026. Congratulations to Ming Cheng!
January 2026
New Team Launch: Our Spatial AI Lab at Inner Mongolia University is officially established! We are looking for motivated students to join our team.
July 2025
Career Started: I am officially joining the College of Computer Science at Inner Mongolia University as a researcher and faculty member.
June 2025
New Role: I successfully defended my Ph.D. thesis at Renmin University of China in June 2025 and obtained a doctoral degree in Computer Science.
February 2025
Paper Accepted: "Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis" accepted by IEEE/CVF CVPR 2025.
December 2024
Attend Conference: I successfully attended the 38th Neural Information Processing Systems (NeurIPS) conference in person in Vancouver, Canada.
October 2024
Academic Service: I am honored to be a volunteer of ACM MultiMedia Conference held at Melbourne Convention Center.
September 2024
Paper Accepted: "Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysis" accepted by NeurIPS 2024.
March 2024
Paper Accepted: "VSFormer: Mining Correlations in Flexible View Set for Multi-view 3D Shape Understanding" accepted by IEEE TVCG 2024.
February 2024
New Journey: I come to an amazing city, Melbourne, Australia and start my one-year academic visit to the Department of Data Science and AI, Monash University, supervised Prof. Jianfei Cai and Dr. Qiuhong Ke.
January 2024
Paper Accepted: "Parameter-efficient Prompt Learning for 3D Point Cloud Understanding" accepted by IEEE ICRA 2024.
July 2023
Scholarship Announcement: I was granted the Joint Doctoral Scholarship of China Scholarship Council (CSC) to support my one-year Ph.D. study abroad.
June 2023
Attend Conference: I successfully attended the IEEE Conference on Robotics and Automation (ICRA) in person in ExCel London, UK.
January 2023
Paper Accepted: "ViPFormer: Efficient Vision-and-Pointcloud Transformer for Unsupervised Pointcloud Understanding" accepted by IEEE ICRA 2023.
September 2022
Paper Accepted: "AirBirds: A Large-scale Challenging Dataset for Bird Strike Prevetion in Real-world Airports" accepted by ACCV 2022.
September 2020
New Role: I am officially started my Ph.D. program at School of Information, Renmin University of China, supervised by Prof. Deying Li and Prof. Yongcai Wang.

Selected Publications

Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis
Hongyu Sun, Qiuhong Ke, Ming Cheng, Yongcai Wang, Deying Li, Chenhui Gou, Jianfei Cai
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025
PDF Code 
@inproceedings{sun25pointcache,
    title={Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis},
    author={Sun, Hongyu and Ke, Qiuhong and Cheng, Ming and Wang, Yongcai and Li, Deying and Gou, Chenhui and Cai, Jianfei},
    booktitle={2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    year={2025},
    pages={1263-1275},
    doi={10.1109/CVPR52734.2025.00126}
}
Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysis
Hongyu Sun, Qiuhong Ke, Yongcai Wang, Wang Chen, Kang Yang, Deying Li, Jianfei Cai
Advances in Neural Information Processing Systems (NeurIPS), 2024
PDF Code 
@inproceedings{sun24point-prc,
    author = {Sun, Hongyu and Ke, Qiuhong and Wang, Yongcai and Chen, Wang and Yang, Kang and Li, Deying and Cai, Jianfei},
    booktitle = {Advances in Neural Information Processing Systems},
    doi = {10.52202/079017-3761},
    editor = {A. Globerson and L. Mackey and D. Belgrave and A. Fan and U. Paquet and J. Tomczak and C. Zhang},
    pages = {118456-118482},
    publisher = {Curran Associates, Inc.},
    title = {Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysis},
    url = {https://proceedings.neurips.cc/paper_files/paper/2024/file/d6938c8e88ef62394d2f4f3fd428e036-Paper-Conference.pdf},
    volume = {37},
    year = {2024}
}
VSFormer: Mining Correlations in Flexible View Set for Multi-view 3D Shape Understanding
Hongyu Sun, Yongcai Wang*, Peng Wang, Haoran Deng, Xudong Cai, Deying Li
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2024
PDF Code 
@ARTICLE{sun25vsformer,
    author={Sun, Hongyu and Wang, Yongcai and Wang, Peng and Deng, Haoran and Cai, Xudong and Li, Deying},
    journal={IEEE Transactions on Visualization and Computer Graphics},
    title={VSFormer: Mining Correlations in Flexible View Set for Multi-View 3D Shape Understanding},
    year={2025},
    volume={31},
    number={4},
    pages={2127-2141},
    keywords={Three-dimensional displays;Shape;Correlation;Feature extraction;Solid modeling;Transformers;Aggregates;Multi-view 3D shape recognition and retrieval;multi-view 3D shape analysis;view set;attention mechanism},
    doi={10.1109/TVCG.2024.3381152}
}
Parameter-efficient Prompt Learning for 3D Point Cloud Understanding
Hongyu Sun, Yongcai Wang, Wang Chen, Haoran Deng, Deying Li
IEEE International Conference on Robotics and Automation (ICRA), 2024
PDF Code 
@INPROCEEDINGS{sun24ppt3d,
    author={Sun, Hongyu and Wang, Yongcai and Chen, Wang and Deng, Haoran and Li, Deying},
    booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)},
    title={Parameter-efficient Prompt Learning for 3D Point Cloud Understanding},
    year={2024},
    pages={9478-9486},
    keywords={Point cloud compression;Solid modeling;Adaptation models;Three-dimensional displays;Prompt engineering;Task analysis;Robotics and automation},
    doi={10.1109/ICRA57147.2024.10610093}
}
ViPFormer: Efficient Vision-and-Pointcloud Transformer for Unsupervised Pointcloud Understanding
Hongyu Sun, Yongcai Wang, Xudong Cai, Xuewei Bai, Deying Li
IEEE International Conference on Robotics and Automation (ICRA), 2023
PDF Code 
@INPROCEEDINGS{sun23vipformer,
    author={Sun, Hongyu and Wang, Yongcai and Cai, Xudong and Bai, Xuewei and Li, Deying},
    booktitle={2023 IEEE International Conference on Robotics and Automation (ICRA)},
    title={ViPFormer: Efficient Vision-and-Pointcloud Transformer for Unsupervised Pointcloud Understanding},
    year={2023},
    pages={7234-7242},
    keywords={Point cloud compression;Solid modeling;Runtime;Three-dimensional displays;Shape;Semantic segmentation;Manuals},
    doi={10.1109/ICRA48891.2023.10160658}
}

Site Statistics

Global Visitor Map

Tracking via Google Analytics | Real-time visitor statistics