Conference paper

HRADIŠ Michal, KOTERA Jan, ZEMČÍK Pavel and ŠROUBEK Filip. Convolutional Neural Networks for Direct Text Deblurring. In: Proceedings of BMVC 2015. Swansea: The British Machine Vision Association and Society for Pattern Recognition, 2015, pp. 1-13. ISBN 1-901725-53-7. Available from: http://www.bmva.org/bmvc/2015/papers/paper006/index.html
Publication language:english
Original title:Convolutional Neural Networks for Direct Text Deblurring
Title (cs):Rekonstrukce rozmazaného textu pomocí konvolučních neuronových sítí
Pages:1-13
Proceedings:Proceedings of BMVC 2015
Conference:British Machine Vision Conference (BMVC) 2015
Place:Swansea, GB
Year:2015
URL:http://www.bmva.org/bmvc/2015/papers/paper006/index.html
ISBN:1-901725-53-7
Publisher:The British Machine Vision Association and Society for Pattern Recognition
Files: 
+Type Name Title Size Last modified
iconhradis15CNNdeblurring.pdf3,52 MB2015-08-17 16:11:25
iconhradis15CNNdeblurring_abstract.pdf637 KB2015-08-17 16:11:29
iconHradis2015_CNN_deblurring.pptx8,79 MB2015-09-08 13:16:19
iconhradis15CNNdeblurring_suplement.zip26,6 MB2015-08-17 16:11:36
^ Select all
With selected:
Keywords
convolutional neural networks, blind deconvolution, image restoration, deblurring, CNN, neural networks, deep learning
Annotation
In this work we address the problem of blind deconvolution and denoising. We focus on restoration of text documents and we show that this type of highly structured data can be successfully restored by a convolutional neural network. The networks are trained to reconstruct high-quality images directly from blurry inputs without assuming any specific blur and noise models. We demonstrate the performance of the convolutional networks on a large set of text documents and on a combination of realistic de-focus and camera shake blur kernels. On this artificial data, the convolutional networks significantly outperform existing blind deconvolution methods, including those optimized for text, in terms of image quality and OCR accuracy. In fact, the networks outperform even state-of-the-art non-blind methods for anything but the lowest noise levels. The approach is validated on real photos taken by various devices. 
Further information including test data and trained networks can be found on the [PROJECT PAGE].

[img]
BibTeX:
@INPROCEEDINGS{
   author = {Michal Hradi{\v{s}} and Jan Kotera and Pavel
	Zem{\v{c}}{\'{i}}k and Filip {\v{S}}roubek},
   title = {Convolutional Neural Networks for Direct Text Deblurring},
   pages = {1--13},
   booktitle = {Proceedings of BMVC 2015},
   year = {2015},
   location = {Swansea, GB},
   publisher = {The British Machine Vision Association and Society for
	Pattern Recognition},
   ISBN = {1-901725-53-7},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=10922}
}

Your IPv4 address: 54.221.76.68
Switch to IPv6 connection

DNSSEC [dnssec]