What is awesome-guided-image-restoration?

This repository is not a piece of software you install or run. It is a curated reading list, sometimes called an awesome list, that collects academic papers, open source code links, and datasets on a research topic called guided or multi-modal image restoration. That field studies how to improve one type of image, such as a low resolution depth photo, by using information from a second related image, such as a normal color photo of the same scene, to fill in missing detail or remove noise. The list is organized by task, covering things like combining several restoration jobs into one all in one model, boosting the resolution of images using a second guiding image, removing noise, brightening low light photos, removing blur, removing haze or rain, and filling in missing or damaged parts of an image. Within resolution boosting, it further breaks things down by the type of guiding image used, such as color photos guiding depth sensors, thermal cameras, hyperspectral sensors, satellite imagery, or medical MRI scans. Each entry in the tables lists the paper's publication year, title, a short name for the method, where it was published, a link to the authors' code if it exists, and a few keywords describing the technique. A second major section collects datasets used for training and testing these methods, again organized by task and split between artificially generated and real world data. The README says the project is continuously updated and welcomes contributions through GitHub issues. There is no installation or usage instructions because there is nothing to run: the value of the repository is purely as an organized reference point for researchers and developers who want to find existing papers, code, and datasets in this specialized area of computer vision. The full README is longer than what was shown.