All One Needs to Know about Priors for Deep Image Restoration and Enhancement: A Survey

Abstract

Image restoration and enhancement is a process of improving the image quality by removing degradations, such as noise, blur, and resolution degradation. Deep learning (DL) has recently been applied to image restoration and enhancement. Due to its ill-posed property, plenty of works have explored priors to facilitate training deep neural networks (DNNs). However, the importance of priors has not been systematically studied and analyzed by far in the research community. Therefore, this paper serves as the first study that provides a comprehensive overview of recent advancements of priors for deep image restoration and enhancement. Our work covers five primary contents: (1) A theoretical analysis of priors for deep image restoration and enhancement; (2) A hierarchical and structural taxonomy of priors commonly used in the DL-based methods; (3) An insightful discussion on each prior regarding its principle, potential, and applications; (4) A summary of crucial problems by highlighting the potential future directions to spark more research in the community; ( 5) An open-source repository that provides a taxonomy of all mentioned works and code links.

Learning from History: Task-agnostic Model Contrastive Learning for Image Restoration

“…Image restoration, aiming at recovering a high-quality image from the degraded one, is a fundamental problem in the fields of image processing and computer vision (Yang et al 2021;Wang, Chen, and Hoi 2021a;Jiang et al 2023). Deep learning approaches have made considerable advancements in image restoration, while there are still challenges due to its ill-posed nature (Wang, Chen, and Hoi 2021b;Lu et al 2023). The success of the self-supervised learning paradigm for high-level tasks, especially those using contrastive learning methods, has drawn great attention (Gui et al 2023a).…”