JPEG is one of the most commonly used standards among lossy image compression methods. However, JPEG compression inevitably introduces various kinds of artifacts, especially at high compression rates, which could greatly affect the Quality of Experience (QoE). Recently, convolutional neural network (CNN) based methods have shown excellent performance for removing the JPEG artifacts. Lots of efforts have been made to deepen the CNNs and extract deeper features, while relatively few works pay attention to the receptive field of the network. In this paper, we illustrate that the quality of output images can be significantly improved by enlarging the receptive fields in many cases. One step further, we propose a Dual-domain Multi-scale CNN (DMCNN) to take full advantage of redundancies on both the pixel and DCT domains. Experiments show that DMCNN sets a new state-of-the-art for the task of JPEG artifact removal.
@article{arxiv.1806.03275,
title = {DMCNN: Dual-Domain Multi-Scale Convolutional Neural Network for Compression Artifacts Removal},
author = {Xiaoshuai Zhang and Wenhan Yang and Yueyu Hu and Jiaying Liu},
journal= {arXiv preprint arXiv:1806.03275},
year = {2018}
}