🧑‍🎓 About Me

Hello! I’m Xincheng Yao (姚欣成), a Ph.D. candidate at Shanghai Jiao Tong University (SJTU), where I’m working under the guidance of Prof. Chongyang Zhang.

I have received the B.S. degree in information engineering from Shanghai Jiao Tong University, China, in 2021. And I’m currently working toward the Ph.D. degree in information and communication engineering.

My academic interests are about Deep Learning, Computer Vision and Anomaly Detection, with a specific focus on Image and Multi-Modal Anomaly Detection. Recently, my research has mainly been centered on how to design more general anomaly detection models, including Multi-Class Anomaly Detection, Zero/Few-shot Anomaly Detection, Class-Agnostic/Generalizable Anomaly Detection, and also the application of large vision language models in anomaly detection tasks. In the anomaly detetion field, I have published multiple papers at the top CV and AI conferences, including CVPR, NeurIPS, ICCV, ECCV, and AAAI.

If you find my research intriguing, please don’t hesitate to contact with me by my email i-dover@sjtu.edu.cn. I welcome any inquries or discussions regarding my work! 😊

🔥 News

• 2024.09: 👏👏👏 Our paper ResAD: A Simple Framework for Class-Generalizable Anomaly Detection, which can address Class-Generalizable Anomaly Detection, is accpected by NuerIPS 2024 as a spotlight paper.

• 2024.09: 👏👏👏 Our paper collected a diverse Cloth Printing Dataset for class-agnostic anomaly detection is accpected by PRCV 2024.

• 2024.07: 👏👏👏 Our paper Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detection, which can address Multi-Class Anomaly Detection, is accpected by ECCV 2024.

• 2023.07: 👏👏👏 Our paper Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detection, which proposed a novel Multi-View Anomaly Detection Framework, is accpected by ICCV 2023.

• 2023.06: 👏👏👏 Our paper CKT: Cross-Image Knowledge Transfer for Texture Anomaly Detection, which can achieve effective Knowledge Transfer among Normal Images, is accpected by ICIP 2023.

• 2023.03: 👏👏👏 Our paper Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detection, which can address Few-Shot Supervised Anomaly Detection, is accpected by CVPR 2023.

• 2022.12: 👏👏👏 Our paper One-for-All: Proposal Masked Cross-Class Anomaly Detection, which can address Multi-Class and Cross-Class Anomaly Detection, is accpected by AAAI 2023.

📃 Publications

First-Authored Peer-Reviewed Publications

Other Peer-Reviewed Publications

First-Authored Peer-Reviewed Publications

1. ResAD: A Simple Framework for Class-Generalizable Anomaly Detection [Paper] [Code].

   Xincheng Yao, Zixin Chen, Chao Gao, Guangtao Zhai, Chongyang Zhang*, The Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS), 2024.

2. Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detection [Paper] [Code].

   Xincheng Yao, Ruoqi Li, Zefeng Qian, Lu Wang, Chongyang Zhang*, The 18th European Conference on Computer Vision (ECCV), 2024.

3. Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detection [Paper] [Code].

   Xincheng Yao, Ruoqi Li, Zefeng Qian, Yan Luo, Chongyang Zhang*, The International Conference on Computer Vision (ICCV), 2023.

4. Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detection [Paper] [Code].

   Xincheng Yao, Ruoqi Li, Jing Zhang, Jun Sun, Chongyang Zhang*, The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023.

5. One-for-All: Proposal Masked Cross-Class Anomaly Detection [Paper] [Code].

   Xincheng Yao, Chongyang Zhang*, Ruoqi Li, Jun Sun, Zhenyu Liu, The 37th Annual AAAI Conference on Artificial Intelligence (AAAI), 2023.

6. Multi-Scale Feature Distillation for Anomaly Detection [Paper].

   Xincheng Yao, Ruoqi Li, Chongyang Zhang*, Kefeng Huang, Kaiyu Sun, The 27th International Conference on Mechatronics and Machine Vision in Practice (M2VIP), 2021.

Other Peer-Reviewed Publications

1. CKT: Cross-Image Knowledge Transfer for Texture Anomaly Detection [Paper].

   Zixin Chen, Xincheng Yao, Zhenyu Liu, Baozhu Zhang, Chongyang Zhang*, The 2023 IEEE International Conference on Image Processing (ICIP), 2023.

2. Enhanced Anomaly Detection Using Spatial-Alignment and Multi-scale Fusion [Paper] [Code].

   Keming Jiao, Xincheng Yao, Lu Wang, Baozhu Zhang, Zhenyu Liu, Chongyang Zhang*, The Chinese Conference on Parttern Recognition and Computer Vision (PRCV), 2024.

🥇 Selected Awards

• National Scholarship for Postgraduates (the highest scholarship for Ph.D.), 2023.09.

• First-class Scholarship for Postgraduates, SJTU, 2022.09, 2023.09, 2024.09.

• National Encouragement Scholarship, 2018.09, 2019.09, 2020.09.

📝 Academic Service

• Conference Reviewer, CVPR, ECCV, ACM MM, etc.

• Journal Reviewer, Engineering Applications of Artificial Intelligence, IEEE Trans. Neural Networks Learn. Syst., IEEE Trans. Circuits Syst. Video Technol., etc.

🎓 Eduacations

• 2021.09 - present, Shanghai Jiao Tong University

  School of Electronic Information and Electrical Engineering

  Ph.D. Candidate in Information and Communication Engineering     Advisor: Chongyang Zhang

• 2017.09 - 2021.06, Shanghai Jiao Tong University

  School of Electronic Information and Electrical Engineering

  B.S. in Information Engineering