SuperInpaint: Learning Detail-Enhanced Attentional Implicit Representation for Super-resolutional Image Inpainting

TL;DR

SuperInpaint introduces a novel single-model approach that combines detail-enhanced implicit representation and attention mechanisms to effectively reconstruct high-resolution images with missing regions, surpassing existing methods.

Contribution

The paper proposes DEAR, a new detail-enhanced attentional implicit representation, enabling high-quality super-resolution inpainting with a single model, addressing limitations of prior combined methods.

Findings

Outperforms existing methods significantly on four metrics.

Extends datasets for the new SuperInpaint task.

Demonstrates effectiveness of detail-enhanced semantic embeddings.

Abstract

In this work, we introduce a challenging image restoration task, referred to as SuperInpaint, which aims to reconstruct missing regions in low-resolution images and generate completed images with arbitrarily higher resolutions. We have found that this task cannot be effectively addressed by stacking state-of-the-art super-resolution and image inpainting methods as they amplify each other's flaws, leading to noticeable artifacts. To overcome these limitations, we propose the detail-enhanced attentional implicit representation (DEAR) that can achieve SuperInpaint with a single model, resulting in high-quality completed images with arbitrary resolutions. Specifically, we use a deep convolutional network to extract the latent embedding of an input image and then enhance the high-frequency components of the latent embedding via an adaptive high-pass filter. This leads to detail-enhanced semantic embedding. We further feed the semantic embedding into an unmask-attentional module that suppresses embeddings from ineffective masked pixels. Additionally, we extract a pixel-wise importance map that indicates which pixels should be used for image reconstruction. Given the coordinates of a pixel we want to reconstruct, we first collect its neighboring pixels in the input image and extract their detail-enhanced semantic embeddings, unmask-attentional semantic embeddings, importance values, and spatial distances to the desired pixel. Then, we feed all the above terms into an implicit representation and generate the color of the specified pixel. To evaluate our method, we extend three existing datasets for this new task and build 18 meaningful baselines using SOTA inpainting and super-resolution methods. Extensive experimental results demonstrate that our method outperforms all existing methods by a significant margin on four widely used metrics.