Deep Artifact-Free Residual Network for Single Image Super-Resolution

TL;DR

This paper introduces a deep residual network for single image super-resolution that effectively extracts high-frequency details while avoiding artifacts, leading to improved image quality.

Contribution

The proposed Deep Artifact-Free Residual network uniquely combines residual learning with ground-truth guidance and a two-step training process for artifact-free high-frequency extraction.

Findings

Achieves superior quantitative super-resolution results.

Produces higher quality images both quantitatively and qualitatively.

Effectively reduces artifacts in reconstructed images.

Abstract

Recently, convolutional neural networks have shown promising performance for single-image super-resolution. In this paper, we propose Deep Artifact-Free Residual (DAFR) network which uses the merits of both residual learning and usage of ground-truth image as target. Our framework uses a deep model to extract the high-frequency information which is necessary for high-quality image reconstruction. We use a skip-connection to feed the low-resolution image to the network before the image reconstruction. In this way, we are able to use the ground-truth images as target and avoid misleading the network due to artifacts in difference image. In order to extract clean high-frequency information, we train the network in two steps. The first step is a traditional residual learning which uses the difference image as target. Then, the trained parameters of this step are transferred to the main training in the second step. Our experimental results show that the proposed method achieves better quantitative and qualitative image quality compared to the existing methods.