Image Inpainting with Edge-guided Learnable Bidirectional Attention Maps

TL;DR

This paper introduces Edge-LBAM, a novel edge-guided learnable attention mechanism for image inpainting that effectively handles irregular holes and preserves image structure, outperforming existing methods.

Contribution

The paper proposes a learnable attention map module and a multi-scale edge completion network to improve inpainting quality by better structural guidance and adaptive mask updating.

Findings

Outperforms state-of-the-art methods in qualitative metrics.

Effectively preserves image structure and coherence.

Reduces color discrepancy and blurriness in inpainted images.

Abstract

For image inpainting, the convolutional neural networks (CNN) in previous methods often adopt standard convolutional operator, which treats valid pixels and holes indistinguishably. As a result, they are limited in handling irregular holes and tend to produce color-discrepant and blurry inpainting result. Partial convolution (PConv) copes with this issue by conducting masked convolution and feature re-normalization conditioned only on valid pixels, but the mask-updating is handcrafted and independent with image structural information. In this paper, we present an edge-guided learnable bidirectional attention map (Edge-LBAM) for improving image inpainting of irregular holes with several distinct merits. Instead of using a hard 0-1 mask, a learnable attention map module is introduced for learning feature re-normalization and mask-updating in an end-to-end manner. Learnable reverse attention maps are further proposed in the decoder for emphasizing on filling in unknown pixels instead of reconstructing all pixels. Motivated by that the filling-in order is crucial to inpainting results and largely depends on image structures in exemplar-based methods, we further suggest a multi-scale edge completion network to predict coherent edges. Our Edge-LBAM method contains dual procedures,including structure-aware mask-updating guided by predict edges and attention maps generated by masks for feature re-normalization.Extensive experiments show that our Edge-LBAM is effective in generating coherent image structures and preventing color discrepancy and blurriness, and performs favorably against the state-of-the-art methods in terms of qualitative metrics and visual quality.