One-Shot Image Restoration
This work addresses the challenge of domain sensitivity and resource-intensive training in image restoration for applications like real-time processing, though it is incremental as it builds on existing patch-based methods.
The paper tackles the problem of high computational and data demands in supervised image restoration by proposing a patch-based learning framework that requires only a single image pair for training, achieving significant improvements in sample efficiency, generalization, and time complexity for deblurring and super-resolution tasks.
Image restoration, or inverse problems in image processing, has long been 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). In addition, 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 be leveraged for future real-time applications, and applied to other signals and modalities.