报告题目：Image recovery：Interaction between regularization
报 告 人：纪辉 副教授 新加坡国立大学数学系
Image recovery is about restoring clear images from degraded ones. It usually requires solving a challenging ill-posed inverse problem. Regularization has been one main tool for image recovery by imposing certain prior on clear images via certain variational form. In this talk, we will discuss how regularization methods and machine learning techniques can interact each other to achieve start-of-the-art performance. Two applications are discussed in this talk to show the benefit of such interaction. One is how Bayesian learning can be introduced in the existing regularization methods for blind motion deblurring of natural images. The other is how L0-norm rel ating regularization can lead a collaborative deep learning method for super-resolving degraded text images.
Dr. Hui Ji received his Ph.D. degree in Computer Science from the University of Maryland at College Park in 2006, and joined National University of Singapore (NUS) ever since. Currently, he is an associate professor in the department of mathematics and is the director of Centre for Wavelets, Approximation and Information Processing. His research interests include computational harmonic analysis, computational vision, and machine learning.