Xiangming Wang

About

I am a Ph.D. student in the Department of Computer Science at Harbin Institute of Technology (Shenzhen), advised during my doctoral studies from 2024 to 2029. My research focuses on multimodal representation learning for image restoration and computational imaging, with an emphasis on vision–language guidance, multi-sensor fusion, and optimization-based deep models.

I am particularly interested in methods that remain structurally faithful under complex degradations while remaining practical for real-world imaging pipelines. I am currently a research intern at Y-Lab, OPPO Research Institute, working on unified denoising and demosaicing for mobile imaging.

Research Interests

• Vision–language image restoration — degradation-aware recovery guided by language priors, including deep unfolding and generative formulations
• Multimodal sensing and fusion — multi-sensor RAW imaging and hyperspectral reconstruction with cross-modal consistency
• Optimization-inspired learning — integrating convex / non-convex optimization principles with diffusion and autoregressive generative models

Education

Experience

Publications

1. [1] Haijin Zeng*, Xiangming Wang*, Yongyong Chen, Jingyong Su, and Jie Liu. “Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. Equal contribution
2. [2] Xiangming Wang, Haijin Zeng, Benteng Sun, Jiezhang Cao, Kai Zhang, Qiangqiang Shen, and Yongyong Chen. “Deep LoRA-Unfolding Networks for Image Restoration.” IEEE Transactions on Image Processing, 2026.
3. [3] Xiangming Wang, Haijin Zeng, Jiaoyang Chen, Sheng Liu, Yongyong Chen, and Guoqing Chao. “OTLRM: Orthogonal Learning-based Low-Rank Metric for Multi-Dimensional Inverse Problems.” Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2025.
4. [4] Xiangming Wang, Benteng Sun, Yungeng Liu, Haijin Zeng, Yongyong Chen, Jingyong Su, and Jie Liu. “HSI-VAR: Rethinking Hyperspectral Restoration through Spatial-Spectral Visual Autoregression.” Under review.
5. [5] Jie Liang, Radu Timofte, et al. (incl. Xiangming Wang). “NTIRE 2025 the 2nd Restore Any Image Model (RAIM) in the Wild Challenge.” CVPR Workshops, 2025. 3rd Place
6. [6] Yaqi Wu, Zhihao Fan, Hirotaka Shinozaki, et al. (incl. Xiangming Wang). “MIPI 2025 Challenge on Deblurring for Hybrid EVS Camera: Methods and Results.” ICCV Workshops, 2025.

Awards & Honors

• National First Prize and Challenge Award Champion, National College Student IoT Design Competition
• Sheng Li Scholarship, Harbin Institute of Technology (top 10 students university-wide; CNY 10,000)
• Tat-Seng Chua Scholarship, Harbin Institute of Technology (top 14 students university-wide; CNY 25,000)
• Academic Scholarship, Harbin Institute of Technology (top 5% university-wide; CNY 22,000)

Technical Skills

Modeling

Vision–language models, deep unfolding, diffusion / autoregressive generative models, multimodal fusion

Data

Synthetic data generation, degradation modeling, cross-modal alignment, spatiotemporal consistency evaluation

Engineering

PyTorch, mobile ISP pipeline development, multimodal training pipelines, CUDA acceleration (basic)

Contact

I welcome discussions and collaborations on multimodal learning, vision–language restoration, and practical imaging systems. Please feel free to reach me at xmwang28@gmail.com.

Google Scholar GitHub Curriculum Vitae