Welcome to my homepage! My name is Shan Jia (贾姗). In Mandarin, my name coincidentally mirrors the pronunciation of “Fake Mountain.” What’s even more interesting is that my research is closely related to “fake things”, including the identification and generation of fake images, videos, audio, and news😀.

I am currently a Software Engineer at YouTube, Google, working on misinformation and synthetic media detection. Before joining Google, I was a Research Scientist and Assistant Lab Director in the Media Forensic Lab led by Professor Siwei Lyu at the University at Buffalo (UB), State University of New York.

Research Interests: Digital Media Forensics, Biometrics, and Computer Vision, etc.

🔈 News

  • 2024.10: 🎉🎉 I’m starting a new role as a Software Engineer in YouTube Trust & Safety team at Google!
  • 2024.09: One paper on multi-aspect text-driven image editing is accepted by NeurIPS 2024 (link).
  • 2024.07: Glad to serve The 20th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS) 2024 as a Publication Chair.
  • 2024.04: Glad to be featured on Buffalo News: link.
  • 2024.04: One paper on using ChatGPT-4V for Deepfake face detection is accepted by CVPR2024 Workshop on Media Forensics (paper, link).
  • 2024.03: Two papers are accepted by ICME 2024 Oral. Congratulations to Soumyya and Yu!
  • 2024.02: Glad to give a talk on Deepfake video detection, invited by Reality Defender.
  • 2024.01: I was interviewed by Futurum Careers, UK together with Prof. Siwei Lyu (link1, link2).
  • 2024.01: One paper on uncovering text-image inconsistency is accepted by ICLR 2024 (link, video).
  • 2023.10: One paper on improving fairness in deepfake detection is accepted by WACV 2024 (Code, news report by UBNow).
  • 2023.09: One paper on audio-visual deepfake detection is accepted by MIT IEEE Undergraduate Research Technology Conference. Congratulations to Sneha Muppalla (a senior high school student)!
  • 2023.08: One paper on street view imagery analysis is accepted by ISPRS Journal of Photogrammetry and Remote Sensing.
  • 2023.08: I’m starting a new position as Assistant Lab Director at UB Media Forensic Lab (UB MDFL)!😊
  • 2023.08: One paper on image forgery detection is accepted by IEEE Transactions on Multimedia.
  • 2023.04: Two papers are accepted by CVPR2023 Workshop on Media Forensics (WMF). Our Autosplice dataset has been released (link)!
  • 2022.11: One paper on street view image inpainting is accepted by ISPRS Journal of Photogrammetry and Remote Sensing (Impact Factor: 12.7).
  • 2022.06: Two papers on Media Forensics are accepted by ICIP 2022.
  • 2022.01: One paper on Image-text De-contextualization Detection is accepted by ICASSP 2022.
  • 2021.12: I gave a keynote speech on Deepfakes at Open Media Forensics Challenge (OpenMFC) 2020-2021 Workshop.
  • 2021.10: I joined UC Berkeley InforCamp Deepfake Panel as a speaker.
  • 2021.06: I defended my Ph.D. dissertation on face spoofing detection and will join UB as a Post-Doctoral Researcher.
  • 2020.12: I joined 2020 International Graduate Workshop on GeoInformatics, and won the Excellent Presentation Award (1/11).

📝 Publications

ICLR2024
sym

[Exposing Text-Image Inconsistency Using Diffusion Models]

Mingzhen Huang, Shan Jia, Zhou Zhou, Yan Ju, Jialing Cai, Siwei Lyu

The Twelfth International Conference on Learning Representations, ICLR 2024, Paper, Code, Video

TMM23
sym

[GLFF: Global and Local Feature Fusion for AI-synthesized Image Detection]

Yan Ju, Shan Jia, Jialing Cai, Haiying Guan, Siwei Lyu

IEEE Transactions on Multimedia, Paper, Code

WMF@CVPR23
sym

[AutoSplice: A Text-prompt Manipulated Image Dataset for Media Forensics]

Shan Jia, Mingzhen Huang, Zhou Zhou, Yan Ju, Jialing Cai, Siwei Lyu

Workshop on Media Forensics, CVPR2023, Paper, Dataset

ISPRS-J23
sym

[UPDExplainer: an Interpretable Transformer-based Framework for Urban Physical Disorder Detection Using Street View Imagery]

Chuanbo Hu, Shan Jia*, Fan Zhang, Changjiang Xiao, Mindi Ruan, Jacob Thrasher, and Xin Li

ISPRS Journal of Photogrammetry and Remote Sensing, Paper

ICIP22
sym

[Model Attribution of Face-swap Deepfake Videos]

Shan Jia, Xin Li, Siwei Lyu

2022 IEEE International Conference on Image Processing, Paper, Code, Dataset

ISPRS-J22
sym

[A Saliency-Guided Street View Image Inpainting Framework for Efficient Last-Meters Wayfinding]

Chuanbo Hu, Shan Jia*, Fan Zhang, Xin Li

ISPRS Journal of Photogrammetry and Remote Sensing, PDF, Code

TCSVT20
sym

[3D face anti-spoofing with factorized bilinear coding]

Shan Jia, Xin Li, Chuanbo Hu, Guodong Guo, Zhengquan Xu

IEEE Transactions on Circuits and Systems for Video Technology, Paper, Dataset

🏫 Education

🖊 Services

  • Conference Reviewer: WACV, ICCV, CVPR, ECCV, ICME
  • Journal Reviewer: IEEE TIFS, IEEE TMM, IEEE TIP, IEEE TCSVT, IJCV, IEEE TBIOM, PR, ACM TOG