北京雁栖湖应用数学研究院

蔡云峰

Zhigang Jia

2025年04月24日 14:00 至 15:00

A3-4-312

In this talk, a new semantic-guided coarse-to-fine generative image inpainting network (SCF-GMIN) is proposed to solve this challenging problem. The main idea is to construct a new two-stage conditional generative adversarial network to fill the missing regions involving multiple semantic categories by introducing external semantic information. Instead of directly inpainting the corrupted image, we leverage the semantic map (SM) inpainted in stage one to guide the inpainting process of corrupted regions in stage two. Both two stages share a coarse-to-fine network that enhances the visual quality of restorations. In our model, the proposed feature module plays a crucial role by storing semantic information of different categories in the feature space with the help of SM. The image inpainting tasks of semantic maps and corrupted images complement each other in the training process, which promotes a balance between understanding image semantics and maintaining visual consistency.

Jia Zhigang, Professor and Doctoral Supervisor at the School of Mathematics and Statistics, Jiangsu Normal University. In 2023, he was selected as a Young and Middle-aged Academic Leader under the "Qinglan Project" of Jiangsu Higher Education Institutions. Since 2024, he has served as an editorial board member of the acdamic journal Numerical Algorithms. His main research interests lie in numerical linear algebra and image processing. Up to now, he has published more than 40 academic papers in journals such as IEEE Transactions on Image Processing, SIAM Journal on Matrix Analysis and Applications, SIAM Journal on Scientific Computing, and SIAM Journal on Imaging Sciences. He has successively made academic visits to the departments of mathematics of universities such as The University of Manchester, Hong Kong Baptist University and University of Macau.