ABSTRACT
Image inpainting, or the reconstruction of antique and corrupted images, has been around for a long time. It is a functional computer vision research problem with the goal of improving an image quality by removing unwanted details, adding missing elements, and presenting the image in a way that appeals to the human visual system. Image inpainting techniques are frequently employed in image reconstruction, such as in museums that may not be able to afford to engage an expert to repair corrupted paintings, image compression, object removal, and so on. This study provides a detailed survey and comparative assessment of several inpainting techniques ranging from basic inpainting techniques to deep learning inpainting techniques. The utility of these techniques is presented with notable comparisons and evaluated by reviewing the various factors as well as available datasets that researchers might utilise to evaluate their suggested method.
