Blind Image Deblurring - What is the Next Step?

Explore the challenges and recent advancements in blind image deblurring during this 46-minute virtual seminar presented by the Society for Industrial and Applied Mathematics. Join speaker Tieyong Zeng from the Chinese University of Hong Kong as he delves into innovative approaches for estimating latent images and blur kernels simultaneously. Discover the surface-aware strategy based on intrinsic geometrical considerations, which preserves sharp edges in intermediate latent images and improves blur kernel estimation. Learn about the method's superior performance in deblurring text and natural images compared to state-of-the-art techniques. Gain insights into Quaternion-based methods for color image restoration and their extension to blind image deblurring. This talk, part of the 39th Imaging & Inverse Problems (IMAGINE) OneWorld SIAM-IS Virtual Seminar Series, offers valuable knowledge for researchers and practitioners in imaging science and related fields.