⛹️‍♂️Ji-Zhe Zhou (周吉喆)🏊‍♂️, Associate Professor at the School of Computer Science, Sichuan University. I obtained my Ph.D. (early graduation) from the University of Macau in 2021. My primary research areas are the self-supervised learning theories and multimedia content security. I am the first or corresponding author for many top conferences and IEEE transactions, including ICCV, NIPS, ACM MM, ACL, IJCAI, AAAI, etc. My publication recieved a total citations.

周吉喆，四川大学计算机学院副教授，主要擅长大模型安全和自监督学习方向，长期招收AI+CV方向solid background and self-motivated的研究生与本科生（限川大），组内各研究方向详见Supervised Students部分，欢迎联系邮件 jzzhou@scu.edu.cn 。

🔥 News

• 2025.05 Publication: ForensicsHub, a unified benchmark & Codebase for All-Domain (IML, Deepfake detect, Document Tamper detect, AIGC detect) Fake Image Detection and Localization is now established by us and under longterm construction & maintain. ForensicsHub is the first-ever effort constructing a unified (all-for-one) model for detecting all-kinds fake images in real life.
• 2025.04 A&S Service: Appointed as the Program Committee Member for NeurIPS 2025.
• 2025.03 A&S Service: Appointed as the Area Chair for MM 2025.
• 2025.03 Publication:  [CVPR 2025]🎉🎉 Style Quantization for Data-Efficient GAN Training, our new quantization method for general GAN training is accepted by CVPR 2025!
• 2025.01 A&S Service: Appointed as the Editorial Board Member for Journal of Machine Learning Research (JMLR) & Applied Energy
• 2024.12 A&S Service: Appointed as the Associate Editor for 《四川大学学报》&《计算机应用》.
• 2024.12 A&S Service: Appointed as the Program Committee Member for ICML25.
• 2024.12 Publication:  [AAAI 2025]🎉🎉 Mesorch, our new IML backbone model, which adopts a parallel CNN+Transformer structure to simultaneously deal with image semantics and non-semantics, is accepted by AAAI25!
• 2024.12 Publication:  [AAAI 2025]🎉🎉 Sparse-ViT, the very first solution of constructing the non-semantic feature extractor through a self-supervised manner in IML is proposed by us and accepted by AAAI 25!
• 2024.11 Competition:  🎉🎉 We won the
  —2nd Price in Alibaba Tianchi, Global AI Attack & Defense Challenge-Track 1: Safety vaccine injection for large model generated images (阿里天池全球AI攻防挑战赛-赛道一：大模型生图安全疫苗注入), meaning 2/1574 .
  —9th Price in Alibaba Tianchi, Global AI Attack & Defense Challenge-Track 2: Financial certificate manipulation detection (阿里天池全球AI攻防挑战赛-赛道二：AI核身之金融场景凭证篡改检测), meaning 9/1162 .
  —Check My Exclusive Interview from Coverpage NEWS (封面新闻个人专访) and the Team Interview from SINA (新浪现场采访).
• 2024.10 A&S Service: Appointed as the Program Committee Member for CVPR25.
• 2024.09 Publication:  [NeurIPS 2024, Spotlight!(0.9%)]🎉🎉 Our comprehensive benchmark and modular codebase on IML (addressing the model evaluation), IMDL-BenCo, is now fully released and under long-term maintenance and construction!
• 2024.09 A&S Service: Appointed as the Program Committee Member for ICLR25.
• 2024.09 Competition:  🎉🎉 We won the 1st Price in Tecent AI Arena Competition (腾讯开悟比赛多智能体博弈赛道), meaning 1/2000+ . Both of our teams are advanced to the grand finals, check Award List for details.
• 2024.07 Publication:  [MM 2024, Oral(3%)]🎉🎉 We raise the generative contrstive approach to address the person identification through chest bones, details are in Neural Boneprint.
• 2024.07 Publication:  [MM 2024, Oral(3%)]🎉🎉 We propose to revise the inexacts while restoring images through Diffusion models, details are in DPPS.
• 2024.06 A&S Service: Appointed as the Program Committee Member for AAAI25.
• 2024.06 Publication:  🎉🎉 Our latest framework and benchmark model on private object detection, PrivacyGuard, is now online; the code and training dataset will soon be available!
• 2024.04 A&S Service: Appointed as the Program Committee Member for NIPS24.
• 2024.01 A&S Service: Appointed as the Area Chair for MM24.
• 2023.12 Publication:  🎉🎉 Our latest benchmark model on IML, IML-ViT, is now fully outsourced!

📝 Selected Publications

Image Generation

• TIFS Personal privacy protection via irrelevant faces tracking and pixelation in video live streaming, Jizhe Zhou, Chi-Man Pun.
• ACM MM 2020 Privacy-sensitive objects pixelation for live video streaming, Jizhe Zhou, Chi-Man Pun, Yu Tong.
• TNNLS Shunting at Arbitrary Feature Levels via Spatial Disentanglement: Toward Selective Image Translation, Jian Wang, Xue Yang, Yuxin Tian, Jizhe Zhou, Jiancheng Lv.

Anti-Generation

• ICCV 2023 (oral) Pre-training-free Image Manipulation Localization through Non-Mutually Exclusive Contrastive Learning, Jizhe Zhou, Xiaochen Ma, Xia Du, Ahmed Y Alhammadi, Wentao Feng,
• To Appear IML-ViT: Image Manipulation Localization by Vision Transformer. Xiaochen Ma, Bo Du, Zhuohang Jiang, Ahmed Y. Al Hammadi, Jizhe Zhou,
• ACM MM 2023 (oral) M2ATS: A Real-world Multimodal Air Traffic Situation Benchmark Dataset and Beyond, Dongyue Guo, Yi Lin, Zheng Zhang, Jizhe Zhou.

Reasoning under DNN

• ACL 2022 Word Segmentation by Separation Inference for East Asian Languages, Yu Tong, Jingzhi Guo, Jizhe Zhou, Ge Chen, Guokai Zheng.
• IJCAI 2022 Abstract rule learning for paraphrase generation, Xianggen Liu, Wenqiang Lei, Jiancheng Lv, Jizhe Zhou.
• TASLP Separation inference: A unified framework for word segmentation in east Asian languages, Yu Tong, Jingzhi Guo, Jizhe Zhou.
• TASLP Towards Recognition for Radio-echo Speech in Air Traffic Control: Dataset and a Contrastive Learning Approach. Yi Lin, Qingyang Wang, Xincheng Yu, Zichen Zhang, Dongyue Guo, Jizhe Zhou.

📖 Educations

• 2013, Bsc., Software Engineering, FST, University of Macau, Macau SAR.
• 2018, Msc., Software Engineering, FST, University of Macau, Macau SAR.
• 2021, Ph.D., Computer and Information Science, FST, University of Macau, Macau SAR.
• 2016 - 2018, Student Research Assistant, CSE, Chinese University of Hong Kong, Hong Kong SAR.
• 2019 - 2020, Student Research Assistant, Kakao Track, Yonsei University, Korean.

💬 Invited Talks

“A Song of Ice and Fire: Image Tampering Techniques in the Age of Large Generative Models”, PRML 2023, Urumqi, China.
“如何为直播中的人脸自动打码”, Academic Salon (a.k.a Job Talk), Sichuan Unviersity 2021.
“Multi-Object Online-Tracking”, University ML Camp (UMLC2019), Jeju, Korean.

📝 Academic Services

• Area Chair: ACM MM 2023
• Publicity Chair: CSAI 2023
• Technical Program Committee: PRML 2023
• Program Committee: AAAI 2021 & 2022, WWW 2021, ICLR 2022, ACM MM 2021 & 2022 & 2023, TCSVT, TIP, TMM, TNNLS, etc..

📝 Supervised Students (2022-now)

• Image Manipulation Localization

  1. Xiaochen Ma, Research Assistant at Sichuan University [Now Ph.D student in MBZUAI, UAE] (Backbone Architecture for IML).
  2. Bo Du, Graduate Student in CS, Sichuan University (IML by Diffusion)
  3. Xinyu Yang, Undergraduate Student in CS, Sichuan University (Data Sythensize for IML)
  4. Lei Su, Master Student in CS, Sichuan University (Anti-AIGC)
  5. Mingxi Guo, Ph.D. Candidate in CIS, University of Macau (Anti-AIGC)

• Reasoning under DNN Architecture

  1. Yu Tong, Lecture, Department of Computer Science and Technology, College of Engineering, Shantou Univerisity (Separation Inference)
  2. Ruiyang Ni, Master Student in CS, Sichuan University (Graph Inference)
  3. Bingkui Tong, Undergraduate Student in CS, Sichuan University (Graph Inference)
  4. Pengsen Liu, Ph.D Student in Lamda Lab, Nanjing University (Game AI)
  5. Zeyu Lei, Undergraduate Student in CS [Now exchange student in University of Macau], Honor College, Sichuan University (Fine-grained Classification)

• Image Generation and Industrial Detection

  1. Jian Wang, Ph.D. Candidate in CS, Sichuan University (Image-to-Image Translation)
  2. Jian Zhang, Master Student in CS, Sichuan University (Defect Detection)
  3. Zhuohang Jiang, Undergraduate Student in CS [Now Ph.D student in PolyU, HongKong], Sichuan University (Deblur \& Image Privacy )
  4. Xuekang Zhu, Master Student in CS, Sichuan University (Game Avatar Generation)