Název:	Compression artifacts removal using convolutional neural networks
Autoři:	Svoboda, Pavel Hradiš, Michal Bařina, David Zemčík, Pavel
Citace zdrojového dokumentu:	Journal of WSCG. 2016, vol. 24, no. 2, p. 63-72.
Datum vydání:	2016
Nakladatel:	Václav Skala - UNION Agency
Typ dokumentu:	článek article
URI:	http://wscg.zcu.cz/WSCG2016/!_2016_Journal_WSCG-No-2.pdf http://hdl.handle.net/11025/21649
ISSN:	1213-6972 (print) 1213-6980 (CD-ROM) 1213-6964 (on-line)
Klíčová slova:	hluboké učení konvoluční neuronová síť JPEG
Klíčová slova v dalším jazyce:	deep learning convolutional neural networks JPEG
Abstrakt v dalším jazyce:	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.
Práva:	© Václav Skala - UNION Agency
Vyskytuje se v kolekcích:	Volume 24, Number 2 (2016)

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.