Scale-Equivariant Imaging: Self-Supervised Learning for Image Super-Resolution and Deblurring

Abstract

Self-supervised methods have recently proved to be nearly as effective as supervised ones in various imaging inverse problems, paving the way for learning-based approaches in scientific and medical imaging applications where ground truth data is hard or expensive to obtain. These methods critically rely on invariance to translations and/or rotations of the image distribution to learn from incomplete measurement data alone. However, existing approaches fail to obtain competitive performances in the problems of image super-resolution and deblurring, which play a key role in most imaging systems. In this work, we show that invariance to roto-translations is insufficient to learn from measurements that only contain low-frequency information. Instead, we propose scale-equivariant imaging, a new self-supervised approach that leverages the fact that many image distributions are approximately scale-invariant, enabling the recovery of high-frequency information lost in the measurement process. We demonstrate throughout a series of experiments on real datasets that the proposed method outperforms other self-supervised approaches, and obtains performances on par with fully supervised learning.

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• Bootstrap your own latent: A new approach to self-supervised Learning
• The Perception-Distortion Tradeoff
• Noise2Noise: Learning Image Restoration without Clean Data
• Exploring Simple Siamese Representation Learning
• Generalized SURE for Exponential Families: Applications to Regularization
• Plug-and-Play Methods for Integrating Physical and Learned Models in Computational Imaging
• Deep Scale-spaces: Equivariance Over Scale
• Sensing Theorems for Unsupervised Learning in Linear Inverse Problems
• Equivariant Bootstrapping for Uncertainty Quantification in Imaging Inverse Problems