Unveiling the Canvas: a Comprehensive Survey of Image Inpainting Techniques

Abstract

Image inpainting is a crucial field in computer vision and image processing that focuses on the reconstruction of missing or damaged parts of images in a visually plausible way. This survey provides an in-depth overview of the various techniques and methodologies developed over the years for image inpainting. We categorize the approaches into three primary classes: traditional methods, deep learning-based methods, and hybrid methods. Traditional techniques include diffusion-based, exemplar-based, and patch-based methods. Deep learning approaches encompass convolutional neural networks (CNNs), generative adversarial networks (GANs), and attention mechanisms. Hybrid methods combine traditional and deep learning techniques to leverage the strengths of both. We evaluate these methods based on their performance, computational complexity, and applicability to different types of image inpainting problems. This survey aims to provide a comprehensive understanding of the advancements in image inpainting and identify potential future research directions.