EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
This addresses the issue of blurry or over-smoothed inpainted images for applications in computer vision and image editing, representing a novel method for a known bottleneck.
The paper tackled the problem of image inpainting by developing EdgeConnect, a two-stage adversarial model that generates edges to guide image completion, resulting in improved reconstruction of fine details and outperforming state-of-the-art methods on datasets like CelebA, Places2, and Paris StreetView.
Over the last few years, deep learning techniques have yielded significant improvements in image inpainting. However, many of these techniques fail to reconstruct reasonable structures as they are commonly over-smoothed and/or blurry. This paper develops a new approach for image inpainting that does a better job of reproducing filled regions exhibiting fine details. We propose a two-stage adversarial model EdgeConnect that comprises of an edge generator followed by an image completion network. The edge generator hallucinates edges of the missing region (both regular and irregular) of the image, and the image completion network fills in the missing regions using hallucinated edges as a priori. We evaluate our model end-to-end over the publicly available datasets CelebA, Places2, and Paris StreetView, and show that it outperforms current state-of-the-art techniques quantitatively and qualitatively. Code and models available at: https://github.com/knazeri/edge-connect