|
Hello! My name is Weiyu Sun. My interests are different kinds of AI techniques (e.g., video-based physiological signal analysis) and their practical applications (e.g., combining these AI techniques with practical devices for healthcare), and trustworthy AI (e.g., the security, and reliability issues of DNN models). I earned my B.E. and M.E. degrees in Electornic & Computer Engineering at Nanjing University, where I mainly dealt with healthcare applications (e.g., wearable devices for ECG measurement, health monitoring). After that I was a visiting student of the Hong Kong University of Science and Technology (Guangzhou) and Pennsylvania state university, where I accomplished three complete works (1st author, one published, another two are under review) and several works cooperated with others (2nd author). Besides, I ever became the research assistant of Nanjing University and got invited to be the technology consultant of Samsung. Currently I'm hunting for a Ph.D. position, if you are interested, be free to contact me! Latest news: one (website) of my papers is accepted by ICLR 2024, cheers! |
|
|
I'm interested in computer vision, machine learning, optimization, and image processing. Some of my research is about video analyzing (mainly focus on physiological signal extraction and processing), and the remaining is about trustworthy AI (the backdoor threats on all kinds of DNN models). Representative papers are highlighted. |
|
Weiyu Sun, Xinyu Zhang, Hao Lu, Yingcong Chen, Jinghui Chen, Lin Lu Accepted by ICLR 2024 We propose a bi-level optimization way to generate the effective trigger pattern to enhance the robustness and effectiveness of the backdoor attack in self-supervised learning scenarios. |
|
Weiyu Sun, Xinyu Zhang, Hao Lu, Ying Chen, Yun Ge, Xiaolin Huang, Jie Yuan, Yingcong Chen Accepted by ACM MM 2023 project page / We reveal two characteristic domain conflict issues in the DNN rPPG (remote photoplethysmography) model training, and propose a new training framework (DOHA) to deal with these conflicts and enhance the performance of rPPG models. |
|
Xinyu Zhang, Weiyu Sun, Hao Lu, Ying Chen, Yun Ge, Xiaolin Huang, Jieyuan, Yingcong Chen, under review in IEEE Transactions on Multimedia We propose a self-supervised training strategy to efficiently train the rPPG DNN model. |
|
Yan Zhang, Han Zhou, Kaiyue Chu, Chuanfeng Wu, Yun Ge, Guoping Shan, Jundong Zhou, Jing Cai, Jianhua Jin, Weiyu Sun, Ying Chen, Xiaolin Huang Frontiers in Oncology paper website The setup accuracy plays an extremely important role in the local control of tumors. The purpose of this study is to verify the feasibility of "Sphere-Mask" Optical Positioning System (S-M_OPS) for fast and accurate setup. |
|
|
 |
Academic LifeBS. degree in Electornic Science (2015-2019)MS. degree in Electornic Science (2019-2022) Awards and ExpertiseC++ Programing [93/100] (2018);Numerical Analysis: [98/100] (2018); Matrix Theory: [96/100] (2020) People's Scholarship of China, 2018; Top 10% in China Post-graduate Mathematical Contest in Modeling, 19th, 2022 Two national patents |
 |
InternshipVisiting Scholoar (online), cooperating with Jinghui Chen and Lin Lu from 2022 to 2023 (till now), the main research topic is the trustworthy AI.Works & Publishments |
 |
InternshipVisiting Scholoar (online), cooperating with Yingcong Chen and Hao Lu from 2022 to 2023 (till now), the main research topic is the video processing via computer vision.Works & Publishment |
|
Thanks for you reading! flowers: |