One-Shot Image Restoration

Abstract

Image restoration, or inverse problems in image processing, is an extensively studied topic. In recent years supervised learning approaches have become a popular strategy attempting to tackle this task. Unfortunately, most supervised learning-based methods are highly demanding in terms of computational resources and training data (sample complexity). Moreover, trained models are sensitive to domain changes, such as varying acquisition systems, signal sampling rates, resolution and contrast. In this work, we try to answer a fundamental question: Can supervised learning models generalize well solely by learning from one image or even part of an image? If so, then what is the minimal amount of patches required to achieve acceptable generalization? To this end, we focus on an efficient patch-based learning framework that requires a single image input-output pair for training. Experimental results demonstrate the applicability, robustness and computational efficiency of the proposed approach for supervised image deblurring and super resolution. Our results showcase significant improvement of learning models’ sample efficiency, generalization and time complexity, that can hopefully inspire future deployment of the proposed learning perspective to future real-time applications, and be applied to other signals and modalities.