|
Xiangyong Lu
I am currently a Postdoctoral Researcher at RIKEN-CCS, working with Prof. Rio Yokota. I received my Ph.D. degree from the Graduate School of Information Sciences at Tohoku University, supervised by
Prof. Takayuki Okatani. Previously, I received my master's degree from School of Computer Science and Technology
at Huazhong University of Science and Technology, supervised by Prof. Chen Yu and Prof. Hai Jin.
My current research focuses on deep neural networks (e.g., Multimodal models) unifying vision, language, and reasoning across diverse domains:
- Agentic multimodal learning for structured information understanding and robots in interactive environments,
- Traffic behavior analysis, hazard/event prediction, planning and reasoning,
- Climate analysis (e.g., simulation, prediction, anomaly detection), and landslide analysis, etc.
My long-term research objectives are to advance AI for Science through scientifically grounded multimodal reasoning and to develop resource-efficient MLLMs for robotic systems.
|
|
- [2026.08] 'CMSA' was accepted for publication in IEE TMM.
- [2024.01] One paper 'SBCFormer' was accepted to WACV'24.
- [2023.08] One paper was accepted to KSEM'23.
- [2024.10-2025.03] Part-time research staff, "Project: Research and development of image feature extraction models in multimodal AI applications", RIKEN AIP.
- [2020.04-2024.10] Research assistant, Graduate School of Information Sciences, Tohoku University.
|
|
Cascaded Multi-Scale Attention for Enhanced Multi-Scale Feature Extraction and Interaction with Low-Resolution Images
Xiangyong Lu,
Masanori Suganuma,
Takayuki Okatani
IEEE Transactions on Multimedia, 2026
arXiv /
code
|
|
|
SBCFormer: Lightweight Network Capable of Full-Size ImageNet Classification at 1 FPS on Single Board Computers
Xiangyong Lu,
Masanori Suganuma,
Takayuki Okatani
WACV, 2024
arXiv /
code
|
|
|
Boosting Light-Weight Depth Estimation Via Knowledge Distillation
Junjie Hu, Chenyou Fan, Hualie Jiang, Xiyue Guo, Yuan Gao, Xiangyong Lu, Tin Lun Lam
The 16th International Conference on Knowledge Science, Engineering and Management, 2023
arXiv /
code
|
|
|
Clairvoyance: Intelligent Route Planning for Electric Buses Based on Urban Big Data
Xiangyong Lu, Kaoru Ota, Mianxiong Dong, Chen Yu, Hai Jin
IEEE Transactions on Intelligent Transportation Systems (Review Results: Accept After Minor Revision), 2020
arXiv
|
|
|
Predicting Transportation Carbon Emission with Urban Big Data
Xiangyong Lu, Kaoru Ota, Mianxiong Dong, Chen Yu, Hai Jin
IEEE Transactions on Sustainable Computing , 2017
arXiv
|
|
|
A Method and System for Predicting Public Transportation Demand Based on Urban Big Data
Hai Jin, Chen Yu, Xiangyong Lu, Jun Chen, Liqiang Niu
Chinese Patent , 2017
|
-
[2021-22] Advance Graduate School Pioneering Research Support Project for Doctoral Students, Japan Science and Technology Agency.
-
Conference Reviewer: CVPR24/25/26, NeurIPS26, ICCV25, ECCV24/26, BMVC26, WACV25/27, ACCV24
-
Journal Reviewer: IJCV24/25/26, Pattern Recognition26, IEICE25
-
In previous years, I also reviewed papers on big data mining and intelligent systems for IEEE Transactions on Vehicular Technology, Vehicular Technology Conference, and CPSCom.
|