Hi 🤗!

I am currently a master’s student at the University of Science and Technology of China (Sep. 2022 - Jun. 2025), advised by Prof. S. Kevin Zhou. I am fortunate to be a member of Medical Imaging, Robotics, Analytic Computing & Learning (MIRACLE) Lab. Prior to this, I obtained my Bachelor’s degree in Computer Science from Hefei University of Technology (Sep. 2018 - Jun. 2022).

My recent work focuses on diffusion models for ID-preserving image synthesis and image editing. If you share similar interests, feel free to drop me an email.

My long-term research goal is to develop generative AI systems that possess creativity on par with professional artists and designers.

I enjoy creating beautiful things, such as photography and painting. I am also an illustrator and have over 50,000 followers on Pixiv and DeviantArt.

📣 I am actively seeking for Researcher / Ph.D. opportunities. Please check out my resume here.

🔥 News

  • May. 2024.  🎉🎉 We release Mojo, a training-free method for image editing by manipulating skip connection features. Check out here for details.
  • May. 2024.  🐧🐧 Join Tencent as a research intern, working on ID-preserving image synthesis.
  • Dec. 2023.  🥇🥇 Win the Suzhou Industrial Park Scholarship.
  • Nov. 2023.  🎉🎉 We are excited to release LoCo, a state-of-the-art layout-to-image synthesis method. LoCo is a training-free plugin designed to enhance the compositional capabilities of text-to-image diffusion models. A demo is available here.
  • Jun. 2023.  🎉🎉 Our DiffULD paper is accepted to MICCAI 2023.

📝 Selected Publications

ArXiv
sym

LoCo: Locally Constrained Training-Free Layout-to-Image Synthesis
Under Review
Peiang Zhao, Han Li, Ruiyang Jin, S. Kevin Zhou$^{\ddagger}$

arXiv | project | demo

  • State-of-the-art training-free method for layout-to-image synthesis.
ArXiv
sym

Mojo: Training-Free Text-guided Image Editing via Skip Connection Modulation
Under Review
Peiang Zhao, Han Li, Ruiyang Jin, S. Kevin Zhou$^{\ddagger}$

arXiv | project

  • A plug-and-play approach for text-guided image editing by modulating skip connection features.
MICCAI 2023
sym

DiffULD: Diffusive Universal Lesion Detection
MICCAI 2023
Peiang Zhao, Han Li, Ruiyang Jin, S. Kevin Zhou$^{\ddagger}$

paper | arXiv

  • Proposed a diffusion model for Universal Lesion Detection (ULD).

🥇 Awards and Scholarships

  • Dec. 2023. Suzhou Industrial Park Scholarship.
  • Oct. 2023. 1st Class Scholarship, University of Science and Technology of China.
  • Oct. 2022. 1st Class Scholarship, University of Science and Technology of China.
  • Jun. 2022. Outstanding Graduate, Hefei University of Technology.
  • Aug. 2021. 2nd Prize, National Intelligent Car Race for University Students.
  • Dec. 2020. 2nd Prize, National Intelligent Car Race for University Students.
  • Jun. 2019. 1st Prize, Competition of Embedded Systems of Anhui Province.

📖 Educations

  • Sep. 2022 - Jun. 2025 (expected). Master student at University of Science and Technology of China.
  • Sep. 2018 - Jun. 2022. Bachelor student at Hefei University of Technology.

💻 Internships

  • 2024.05 - Present. Tencent, QQ Image Lab, Shanghai, China.
  • 2020.06 - 2020.08. FISTCO Shanghai R&D Center, Shanghai, China.