A Comprehensive Study of Image Inpainting Techniques with Algorithmic approach
Tammineni Shanmukhaprasanthi, Swarajya Madhuri Rayavarapu, Yenneti Laxmi Lavanya, Gottapu Sasibhushana Rao · 2023
The process of image inpainting is indeed a method that allows you to fill in particular regions of an image using novel content by estimating from surrounding pixels, from other images, or from various sources. The term “image inpainting” refers to a wide range of techniques used to repair visual media, from restoring old photographs and films to erasing undesirable text or objects. When only individual components of a picture are missing, traditional methods can produce high-quality solutions, but they are unable to detect new elements that were not included in the picture. Improvements in the quality of image inpainting have been achieved by the use of Deep Learning techniques, which have made it possible to generate appropriate hole filling and separate objects that aren’t included in a real image. However, there is a great deal of room for development in this regard, especially in the areas of adapting to different image sizes, using free-form masks, making high-resolution textures, using fewer computational resources, and shortening the training process. The comprehensive study carried on various image inpainting techniques and their applications will be useful to the new researchers to explore the possibility of developing new efficient algorithms for further improved image inpainting techniques truer to life.