DAM-GAN : Image Inpainting using Dynamic Attention Map based on Fake Texture Detection

TL;DR

DAM-GAN is a novel image inpainting model that uses dynamic attention maps to detect and reduce fake textures, improving image quality by minimizing artifacts and color inconsistencies.

Contribution

The paper introduces DAM-GAN, a GAN-based inpainting method that dynamically detects fake textures to enhance image synthesis quality.

Findings

Outperforms existing inpainting methods on CelebA-HQ and Places2 datasets.

Effectively reduces pixel artifacts and color inconsistencies.

Demonstrates superior image reconstruction quality.

Abstract

Deep neural advancements have recently brought remarkable image synthesis performance to the field of image inpainting. The adaptation of generative adversarial networks (GAN) in particular has accelerated significant progress in high-quality image reconstruction. However, although many notable GAN-based networks have been proposed for image inpainting, still pixel artifacts or color inconsistency occur in synthesized images during the generation process, which are usually called fake textures. To reduce pixel inconsistency disorder resulted from fake textures, we introduce a GAN-based model using dynamic attention map (DAM-GAN). Our proposed DAM-GAN concentrates on detecting fake texture and products dynamic attention maps to diminish pixel inconsistency from the feature maps in the generator. Evaluation results on CelebA-HQ and Places2 datasets with other image inpainting approaches show the superiority of our network.