IMDL-BenCo: A Comprehensive Benchmark for Image Manipulation Detection & Localization
Advances in Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks Track, 2024
周吉喆,四川大学计算机学院副教授, 四川省千人计划入选者, IEEE TETCI 副主编, J. Imaging 客座主编。 主要研究方向为伪造图像检测(Fake Image Detection)及生成与取证统一(Forensic-Embedded Generation)。 主导开发了 IML-ViT、Sparse-ViT、IMDL-BenCo、ForensicHub、SICA(OpenMMSec)、Defake 等多个开源项目。 更多详情请见 GitHub 与 Google Scholar 主页。
Ji-Zhe Zhou is an Associate Professor at the College of Computer Science, Sichuan University, and a recipient of the Sichuan Provincial Thousand Talents Program. He serves as Associate Editor of IEEE TETCI and Guest Editor of J. Imaging. His research focuses on Fake Image Detection and Forensic-Embedded Generation. He has led the development of several open-source projects, including IML-ViT, Sparse-ViT, IMDL-BenCo, ForensicHub, SICA (OpenMMSec), and Defake. For more details, please visit his GitHub and Google Scholar profiles.
SICA preprint — Can We Build a Monolithic Model for Fake Image Detection? — on arXiv.
Advances in Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks Track, 2024
IEEE/CVF International Conference on Computer Vision (ICCV), pp. 22346–22356, 2023
Advances in Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks Track, 2025
arXiv preprint arXiv:2307.14863, 2023
arXiv preprint arXiv:2602.06676, 2026
Sichuan University, College of Computer Science
四川省千人计划入选者University of Macau
Ph.D. Supervisor: Prof. Chi-Man Pun