About Me

I am an Applied Scientist at Amazon AGI, where I focus on core research and development for foundational generative AI.

I earned my Ph.D. in Computer Science from the UIUC in 2024, where I was fortunate to be advised by Prof. Svetlana Lazebnik. Prior to that, I completed my B.S. in Computer Science at Penn State University.

My research focuses on generative image and video modeling. I am passionate about tackling fundamental challenges like high-fidelity detail preservation and grounding advanced foundation models in practical, real-world applications such as virtual try-on.

News

• Sep 2025 "Unziplora" has been accepted to ICCV 2025!
• Jan 2025 "Street TryOn" is accepted to WACV 2025.

Selected Research

[Nova2]

Amazon Nova 2: Multimodal Reasoning and Generation Models

Amazon AGI

Technical Report 2025 [Paper]

[Unziplora]

Unziplora: Separating content and style from a single image

C. Liu, V. Shah, Aiyu Cui, S. Lazebnik

ICCV 2025 [Project Page] [Paper] [Code]

[Street TryOn]

Street TryOn: Learning In-the-Wild Virtual Try-On from Unpaired Person Images

Aiyu Cui, J. Mahajan, V. Shah, P. Gomathinayagam, C. Liu, S. Lazebnik

WACV 2025 Best Paper @ CVFAD CVPR Workshop 2024 [Project Page] [Paper] [Code]

[DensePose]

Learning Garment DensePose for Robust Warping in Virtual Try-On

Aiyu Cui, S. He, T. Xiang, A. Toisoul

arXiv 2023 [Paper]

[Dressing]

Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-On and Outfit Editing

Aiyu Cui, Daniel McKee, Svetlana Lazebnik

ICCV 2021 Best Paper @ CVFAD CVPR Workshop 2021 [Project Page] [Code] [Paper]

Experience

Education