Abstract

Deep learning has proved to be a powerful tool in many domains, including inverse imaging problems. However, most existing successful deep learning solutions to these inverse problems are based on supervised learning, which requires many ground-truth images for training a deep neural network (DNN). This prerequisite on training datasets limits their applicability in data-limited domains, such as medicine and science. In this talk, we will introduce a series of works on self-supervised learning for solving inverse imaging problems. Our approach teaches a DNN to predict images from their noisy and partial measurements without seeing any related truth image, which is achieved by neuralization of Bayesian inference with DNN-based over-parametrization of images. Surprisingly, our proposed self-supervised method can compete well against supervised learning methods in many real-world imaging tasks.

Speaker

Dr. Ji Hui obtained his Ph.D. in Computer Science from the University of Maryland at College Park in 2006. He currently is an Associate Professor at the Department of Mathematics and serves as the Deputy Director of the Centre for Data Science and Machine Learning at NUS. He serves on the editorial boards of several research journals, including the SIAM Journal on Imaging Sciences. His research interests lie in computational harmonic analysis, computational vision, imaging science, and machine learning.