Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting

By Zili Yi, Qiang Tang, Shekoofeh Azizi, Daesik Jang, Zhan Xu. 2020

349 citations recorded by the source corpus.

Recently data-driven image inpainting methods have made inspiring progress, impacting fundamental image editing tasks such as object removal and damaged image repairing. These methods are more effective than classic approaches, however, due to memory limitations they can only handle low-resolution inputs, typically smaller than 1K. Meanwhile, the resolution of photos captured with mobile devices increases up to 8K. Naive up-sampling of the low-resolution inpainted result can merely yield a large yet blurry result. Whereas, adding a high-frequency residual image onto the large blurry image can generate a sharp result, rich in details and textures. Motivated by this, we propose a Contextual Residual Aggregation (CRA) mechanism that can produce high-frequency residuals for missing contents by weighted aggregating residuals from contextual patches, thus only requiring a low-resolution prediction from the network. Since convolutional layers of the neural network only need to operate on low-resolution inputs and outputs, the cost of memory and computing power is thus well suppressed. Moreover, the need for high-resolution training datasets is alleviated. In our experiments, we train the proposed model on small images with resolutions 512 × 512 and perform inference on high-resolution images, achieving compelling inpainting quality. Our model can inpaint images as large as 8K with considerable hole sizes, which is intractable with previous learning-based approaches. We further elaborate on the light-weight design of the network architecture, achieving real-time performance on 2K images on a GTX 1080 Ti GPU. Codes are available at: https://github. com/Ascend-Huawei/Ascend-Canada/tree/ master/Models/Research_HiFIll_Model

Subfields: Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Generative Models

Citations in the corpus

MaskGIT: Masked Generative Image Transformer (2022, methodology): “We compare MaskGIT against common GAN-based baselines, including DeepFillv2 [52] and HiFill [50], on in-painting with a central 50% ˆ 50% mask, which are evaluated on the Places2 validation set.”
Palette: Image-to-Image Diffusion Models (2021, methodology): “It is important to note that DeepFillv2 and HiFill are not trained on ImageNet, but we report their performance on ImageNet ctest10k primarily for benchmarking purposes.”
MAT: Mask-Aware Transformer for Large Hole Image Inpainting (2022, methodology): “To explicitly model long-range dependencies in inpainting, [61, 65, 66] propose to employ attention modules in the CNN-based generator.”
Aggregated Contextual Transformations for High-Resolution Image Inpainting (2021, methodology): “[7] and HiFill [8] distort the structures in results by unreasonable repeating patterns.”
DiffIR: Efficient Diffusion Model for Image Restoration (2023, background): “Up to now, researchers have carried out CNN’s study with different perspectives and obtained more elaborate network architecture designs and learning schemes, such as residual block [29, 81, 6], GAN [21, 65, 48], attention [86, 66, 11, 72, 71, 68, 73], knowledge distillation [67], and others [26, 19”