Publication Details
Compression Artifacts Removal Using Convolutional Neural Networks
Hradiš Michal, Ing., Ph.D. (DCGM FIT BUT)
Bařina David, Ing., Ph.D. (DCGM FIT BUT)
Zemčík Pavel, prof. Dr. Ing. (DCGM FIT BUT)
Deep learning, Convolutional neural networks, JPEG, Compression artifacts, Deblocking, Deringing
This paper shows that it is possible to train large and deep convolutional neural networks (CNN) for JPEG compression artifacts reduction.
This paper shows that it is possible to train large and deep convolutional neural networks (CNN) for JPEG compression artifacts reduction, and that such networks can provide significantly better reconstruction quality compared to previously used smaller networks as well as to any other state-of-the-art methods. We were able to train networks with 8 layers in a single step and in relatively short time by combining residual learning, skip architecture, and symmetric weight initialization. We provide further insights into convolution networks for JPEG artifact reduction by evaluating three different objectives, generalization with respect to training dataset size, and generalization with respect to JPEG quality level.
@ARTICLE{FITPUB11176,
author = "Pavel Svoboda and Michal Hradi\v{s} and David Ba\v{r}ina and Pavel Zem\v{c}\'{i}k",
title = "Compression Artifacts Removal Using Convolutional Neural Networks",
pages = "63--72",
journal = "Journal of WSCG",
volume = 24,
number = 2,
year = 2016,
ISSN = "1213-6972",
language = "english",
url = "https://www.fit.vut.cz/research/publication/11176"
}