Abstract
Many state-of-the-art image restoration approaches do not scale well to larger images, such as megapixel images common in the consumer segment. Computationally expensive optimization is often the culprit. While effificient alternatives exist, they have not reached the same level of image quality. The goal of this paper is to develop an effective approach to image restoration that offers both computational effificiency and high restoration quality. To that end we propose shrinkage fifields, a random fifield-based architecture that combines the image model and the optimization algorithm in a single unit. The underlying shrinkage operation bears connections to wavelet approaches, but is used here in a random fifield context. Computational effificiency is achieved by construction through the use of convolution and DFT as the core components; high restoration quality is attained through loss-based training of all model parameters and the use of a cascade architecture. Unlike heavily engineered solutions, our learning approach can be adapted easily to different trade-offs between effificiency and image quality. We demonstrate state-of-the-art restoration results with high levels of computational effificiency, and signifificant speedup potential through inherent parallelism