Hi, I'm Duowang Zhu (朱多旺), a Ph.D. student at LIESMARS, Wuhan University.
My research lies at the intersection of deep learning and remote sensing, with a focus on generative models and spatiotemporal modeling for Earth observation.

• Email: zhuduowang@whu.edu.cn
• Google Scholar: Profile

• Generative Foundation Models: Generalization, interpretability, and downstream adaptation in generative diffusion models.
• Multi-Temporal Remote Sensing: Modeling temporal dynamics for tasks including time series reconstruction, change detection & captioning.
• Video & Sequence Modeling: Bridging video understanding techniques with remote sensing applications.

Duowang Zhu, Xiaohu Huang, Haiyan Huang, Hao Zhou, Zhenfeng Shao
CVPR 2025 (🏆 Highlight, Top 3%)
Paper | Code

Proposes a unified video modeling framework for change detection and captioning in remote sensing.

Duowang Zhu, Xiaohu Huang, Haiyan Huang, Qimin Cheng, Zheng Huang, Zhenfeng Shao
Pattern Recognition 2025
Paper | Code

Explores the potential of plain ViT architectures for bi-temporal change detection without complex designs.

Xiaohu Huang^†, Duowang Zhu^†, Hao Wang, Xinggang Wang, Bo Yang, Botao He, Wenyu Liu, Bin Feng
ICCV 2021 (^†Equal Contribution)
Paper | Code

Introduces a temporal modeling framework for gait recognition with context-aware attention.

I have served as a reviewer for journals and conferences in computer vision, artificial intelligence, and remote sensing, including: TGRS, TMM, CVPR, ICCV, ECCV, NeurIPS, etc.