Tianjin University

Xuanpu Zhang

I am currently pursuing a Ph.D. at Tianjin University and expect to graduate in 2029. I received my M.S. degree in computer vision from Tianjin University, advised by Prof. Dan Song, and my B.S. degree in Electronic Information from Hebei University of Technology, where I worked with Prof. Zhenxv Bai.

My research interests lie in visual content generation, with a current focus on video models and world models. Previously, I worked on image generation, image editing, and virtual try-on.

I am currently a research intern at Kling AI Research. Before that, I interned at Kuaishou Technology until July 2026 and at Alibaba in 2024.

Email Google Scholar GitHub

Updates

News

2026.08 Our new paper Dior is available on arXiv, with an official GitHub repository and interactive demo.
2026.07 Our paper DARS was accepted to ACM MM 2026.
2026.02 Our paper LayerBind was accepted to CVPR 2026 as a poster.
2026.02 Our paper GRAG was accepted to CVPR 2026 Findings.
2026.02 Our paper Eevee was accepted to CVPR 2026 Findings.

Background

Experience

Research Internships

Kling AI Research

Research Intern

2026.07 – Present

Kuaishou Technology

Research Intern

2025 – 2026.07

Alibaba

Research Intern

2024

Education

Tianjin University

Ph.D. Student · Expected 2029

M.S. in Computer Vision · 2022–2025

Selected work

Research Papers

2026

DiOR-Light scribble-guided image relighting examples

arXiv 2026

Dior: Drawing the Light of Image via Material-Decoupled Illumination Representation

Xuanpu Zhang, Xuesong Niu, Haoxiang Cao, Ruidong Chen, Jianhao Zeng, Changqian Yu

arXiv preprint, 2026

Paper GitHub Demo

DARS method overview

ACM MM 2026

DARS: Dual-Level Credit Assignment RL with Structured Reasoning for Instruction-Based Image Editing

Haoxiang Cao, Jiajiong Cao, Xuanpu Zhang, Changqian Yu, Chaoqun Wang

ACM International Conference on Multimedia (ACM MM), 2026

Paper

LayerBind results and applications

CVPR 2026 · Poster

Layer-wise Instance Binding for Regional and Occlusion Control in Text-to-Image Diffusion Transformers

Ruidong Chen, Yancheng Bai, Xuanpu Zhang, Jianhao Zeng, Lanjun Wang, Dan Song, Lei Sun, Xiangxiang Chu, Anan Liu

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026 · Poster

Paper Project

GRAG continuous image editing results

CVPR 2026 Findings

Group Relative Attention Guidance for Image Editing

Xuanpu Zhang, Xuesong Niu, Ruidong Chen, Dan Song, Jianhao Zeng, Penghui Du, Haoxiang Cao, Kai Wu, An-an Liu

IEEE/CVF Conference on Computer Vision and Pattern Recognition Findings (CVPR Findings), 2026

Paper Project Code Demo

Eevee project illustration

CVPR 2026 Findings

Eevee: Towards Close-up High-resolution Video-based Virtual Try-on

Jianhao Zeng, Yancheng Bai, Ruidong Chen, Xuanpu Zhang, Lei Sun, Dongyang Jin, Ryan Xu, Nannan Zhang, Dan Song, Xiangxiang Chu

IEEE/CVF Conference on Computer Vision and Pattern Recognition Findings (CVPR Findings), 2026

Paper Code Dataset

PEMF-VTO method overview

IEEE TCE 2026

PEMF-VTO: Point-Enhanced Video Virtual Try-on via Mask-Free Paradigm

Tianyu Chang, Xiaohao Chen, Zhichao Wei, Xuanpu Zhang, Qingguo Chen, Weihua Luo, Peipei Song, Xun Yang

IEEE Transactions on Consumer Electronics (TCE), 2026

Paper Project

2025

BooW-VTON results

CVPR 2025

BooW-VTON: Boosting In-the-Wild Virtual Try-On via Mask-Free Pseudo Data Training

Xuanpu Zhang, Dan Song, Pengxin Zhan, Tianyu Chang, Jianhao Zeng, Qingguo Chen, Weihua Luo, An-An Liu

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

Paper Code

2024

Better Fit virtual try-on results

TCSVT

Better Fit: Accommodate Variations in Clothing Types for Virtual Try-on

Dan Song, Xuanpu Zhang, Jianhao Zeng, Pengxin Zhan, Qingguo Chen, Weihua Luo, An-An Liu

IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

Paper

Image-based virtual try-on survey overview

IJCV 2024

Image-Based Virtual Try-On: A Survey

Dan Song, Xuanpu Zhang, Juan Zhou, Weizhi Nie, Ruofeng Tong, Mohan S. Kankanhalli, An-An Liu

International Journal of Computer Vision (IJCV), 2024

Paper