Faculty of Information Technology, BUT

Publication Details

Feature extraction for efficient image and video segmentation

VOJVODA Jakub and BERAN Vítězslav. Feature extraction for efficient image and video segmentation. In: Proceedings of SCCG 2016. Smolenice: Association for Computing Machinery, 2016, pp. 1-7. ISBN 978-1-4503-3693-2.
Czech title
Extrakce příznaků pro efektivní segmentaci obrazu a videa
Type
conference paper
Language
english
Authors
Vojvoda Jakub, Ing. (FIT BUT)
Beran Vítězslav, Ing., Ph.D. (DCGM FIT BUT)
Keywords
Color/Texture segmentation, Motion segmentation, RGB-D/T segmentation
Abstract
The segmentation of sensory data of various domains is often crucial pre-processing step in many computer vision methods and applications. In this work, we propose a method that leverages the quantization of local features distributions for the depth and the temporal information. Three variants of the segmentation method is designed and evaluated reflecting various data domains: space (color and texture), temporal (motion) and depth domain. Each variant was tested on appropriate dataset showing the usability of designed method for applications like areal-image analysis, hand detection and moving-people detection. The pilot experiments shows the characteristics of the approach and computational costs of designed variants. 
Published
2016
Pages
1-7
Proceedings
Proceedings of SCCG 2016
Conference
Spring Conference on Computer Graphics 2016, Smolenice, SK
ISBN
978-1-4503-3693-2
Publisher
Association for Computing Machinery
Place
Smolenice, SK
DOI
BibTeX
@INPROCEEDINGS{FITPUB11086,
   author = "Jakub Vojvoda and V\'{i}t\v{e}zslav Beran",
   title = "Feature extraction for efficient image and video segmentation",
   pages = "1--7",
   booktitle = "Proceedings of SCCG 2016",
   year = 2016,
   location = "Smolenice, SK",
   publisher = "Association for Computing Machinery",
   ISBN = "978-1-4503-3693-2",
   doi = "10.1145/2948628.2948631",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/11086"
}
Files
Back to top