🌟 About Me
I am currently a research assistant at the Graduate School of Engineering, University of Tokyo. My research focuses on computer vision, deep learning, and graph neural networks, particularly in the context of human motion understanding and representation. I am actively seeking PhD opportunities.
Feel free to contact me if you would like more information or access to the code related to my work.
📝 Publications
An Animation-based Augmentation Approach for Action Recognition from Discontinuous Video
Xingyu Song, Zhan Li, Shi Chen, Xin-Qiang Cai, Kazuyuki Demachi, Proceedings of the 27th European Conference on Artificial Intelligence (ECAI’24), Santiago de Compostela, Spain, Oct. 19-24, 2024 (Oral).
- 4A (Action Animation-based Augmentation Approach) is a pipeline that enhances action recognition by generating animated pose data through 2D pose estimation, Quaternion-based GCN, and Dynamic Skeletal Interpolation, which effectively bridging the gap between virtual and real-world data and achieving superior performance with significantly less data.
Quater-GCN: Enhancing 3D Human Pose Estimation with Orientation and Semi-supervised Training
Xingyu Song, Zhan Li, Shi Chen, Kazuyuki Demachi, Proceedings of the 27th European Conference on Artificial Intelligence (ECAI’24), Santiago de Compostela, Spain, Oct. 19-24, 2024 (Oral).
- Quater-GCN (Q-GCN) is a deep learning model that enhances 3D human pose estimation by incorporating both joint spatial dependencies and bone orientation, utilizing a directed graph convolutional network and a semi-supervised training strategy, resulting in superior performance compared to existing methods.
-
Game Engine Based Data Augmentation with In-game Customization and Modeling for Malicious Behaviors Identification in Nuclear Security, Xingyu Song, Zhan Li, Shi Chen, Kazuyuki Demachi, INMM/ESARDA 2023 Joint Annual Meeting.
-
GTAutoAct: An Automatic Datasets Generation Framework Based on Game Engine Redevelopment for Action Recognition, Xingyu Song, Zhan Li, Shi Chen, Kazuyuki Demachi, arXiv Preprint
-
Data, language and graph-based reasoning methods for identification of human malicious behaviors in nuclear security, Zhan Li, Xingyu Song, Shi Chen, Kazuyuki Demachi, Expert Systems with Applications, Vol.236, 121367.
-
Abnormal Detection in Nuclear Security Videos Based on Label-Specific Autoencoders and Reconstruction Errors Comparison, Zhan Li, Xingyu Song, Shi Chen, Kazuyuki Demachi, Nuclear Engineering and Technology (Forthcoming).
-
Armed Boundary Sabotage: A Case Study of Human Malicious Behaviors Identification with Computer Vision and Explainable Reasoning Methods, Zhan Li, Xingyu Song, Shi Chen, Kazuyuki Demachi, Computers and Electrical Engineering, Vol.121, 109924.
-
Malicious behaviors identification in nuclear security based on visual relationships extraction and knowledge reasoning, Zhan Li, Xingyu Song, Shi Chen, Kazuyuki Demachi, INMM/ESARDA 2023 Joint Annual Meeting.
-
Advancement and Development of Graph-Based Reasoning Method for Human Malicious Behaviors Identification Based on Graph Contrastive Representation Learning, Zhan Li, Xingyu Song, Shi Chen, Kazuyuki Demachi, Neurocomputing (Under review).
📖 Educations
- 2022.04 - 2024.03, M.Eng., Graduate School of Engineering, The University of Tokyo.
- 2021.10 - 2022.03, Research Student, School of Fundamental Science and Engineering, Waseda University.
- 2017.09 - 2021.07, B.Sc., College of Hongshen/College of Computer Science, Chongqing University.
🏆 Honors and Awards
- 2024，SPRING GX Fellowship, The University of Tokyo.
💼 Internships
- 2020.06 – 2020.08, Sichuan Hwadee Information Technology Co., Ltd, Chengdu, China.
- 2019.01 – 2019.03, National key laboratory of Software Development Environment, Beihang University, Beijing, China. Supervised by Prof. Zhiming Zheng.
- 2018.07 – 2018.09, ChinaSoft International, Chongqing, China. CQUHub