Blind image deblurring with unknown boundaries using the alternating direction method of multipliers

摘要: Blind image deblurring (BID) is an ill-posed inverse problem, typically solved by imposing some form of regularization (prior knowledge) on the unknown blur and original image. A recent approach, although not requiring prior knowledge blurring filter, achieves state-of-the-art performance for a wide range real-world BID problems. We propose new version that method, in which both optimization problems with respect to are alternating direction method multipliers (ADMM) - tool has recently sparked much interest solving problems, namely due its modularity speed. Our approach also handles seamlessly realistic case blind boundary conditions. Experiments synthetic real blurred images show competitiveness proposed terms speed restoration quality.