Zhengyu Zhao


Google Scholar




About Me

I am a final-year PhD candidate at Radboud University in the Netherlands, supervised by Prof. Martha Larson. I will be joining CISPA Helmholtz Center for Information Security in Germany as a postdoc, working with Prof. Michael Backes. My general research goal is to develop trustworthy computer vision systems. Much of my experience has concentrated on adversarial image examples. I also have a background in image forensics and scene recognition.


- [09/2021] I have finally submitted my PhD thesis after a long journey of four years.
- [05/2021] I am honored to participate in the Doctoral Consortium of CVPR 2021.
- [08/2020] I am an Outstanding Reviewer of BMVC 2020.
- [07/2020] We are organizing the 3rd edition of Pixel Privacy Task at MediaEval Benchmark 2020. The task will be focused on deceiving the Blind Image Quality Assessment (BIQA) model by adversarially modifying images.


On Success and Simplicity: A Second Look at Transferable Targeted Attacks
Zhengyu Zhao, Zhuoran Liu, Martha Larson
arXiv, 2021 (Abridged version accepted to ICML 2021 Workshop on AdvML)

Pivoting Image-based Profiles Toward Privacy: Inhibiting Malicious Profiling with Adversarial Additions
Zhuoran Liu, Zhengyu Zhao, Martha Larson
ACM Conference on User Modeling, Adaptation and Personalization (UMAP), 2021

Screen Gleaning: A Screen Reading TEMPEST Attack on Mobile Devices
Exploiting an Electromagnetic Side Channel

Zhuoran Liu, Niels Samwel, Léo Weissbart, Zhengyu Zhao, Dirk Lauret, Lejla Batina, Martha Larson
Network and Distributed System Security Symposium (NDSS), 2021
[Code] [Video]

Adversarial Robustness Against Image Color Transformation
within Parametric Filter Space

Zhengyu Zhao, Zhuoran Liu, Martha Larson
Under review by a journal, 2020
[Code] [Video] [Preliminary Version at BMVC 2020]

Towards Large yet Imperceptible Adversarial Image Perturbations
with Perceptual Color Distance

Zhengyu Zhao, Zhuoran Liu, Martha Larson
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
[Code] [Video]

Who's Afraid of Adversarial Queries? The Impact of Image Modifications
on Content-based Image Retrieval

Zhuoran Liu, Zhengyu Zhao, Martha Larson
ACM International Conference on Multimedia Retrieval (ICMR), 2019

From Volcano to Toyshop: Adaptive Discriminative Region Discovery
for Scene Recognition

Zhengyu Zhao, Martha Larson
ACM International Conference on Multimedia (ACM MM), 2018
[Code] [Reproducibility companion paper at ACM MM 2019]

Academic Service

  • Journal reviewer of IEEE Transactions on Information Forensics and Security, Pattern Recognition.
  • Conference reviewer/PC Member of AAAI 2022, BMVC 2020-2021.
  • Task co-organizer of Pixel Privacy (2018-2020) and Multimedia Satellite (2018, 2019) at MediaEval Benchmark.
  • Local organizing team (Poster Session Chair) of ACM MM 2019.

    Honors & Awards

  • Participant of Doctoral Consortium, CVPR 2021
  • Our team ranked top 1% in CVPR 2021 Security AI Challenge: Unrestricted Adversarial Attacks on ImageNet
  • Outstanding Reviewer (55/970), BMVC 2020
  • Student Travel Grant, ACM MM 2018 & 2019

  • Miscellaneous

  • I like conceptually simple yet effective research ideas since I am a bit lazy.
  • I respect Dr. Nicholas Carlini, who dedicates himself to rightful evaluation on adversarial robustness.
  • I love music, and I was ranked among the top 10 in a singing competition for Chinese overseas in the Netherlands.