DocumentCode :
2504693
Title :
Feature Pairs Connected by Lines for Object Recognition
Author :
Awais, Muhammad ; Mikolajczyk, Krystian
Author_Institution :
Centre for Vision, Speech & Signal Process. (CVSSP), Univ. of Surrey, Guildford, UK
fYear :
2010
fDate :
23-26 Aug. 2010
Firstpage :
3093
Lastpage :
3096
Abstract :
In this paper we exploit image edges and segmentation maps to build features for object category recognition. We build a parametric line based image approximation to identify the dominant edge structures. Line ends are used as features described by histograms of gradient orientations. We then form descriptors based on connected line ends to incorporate weak topological constraints which improve their discriminative power. Using point pairs connected by an edge assures higher repeatability than a random pair of points or edges. The results are compared with state-of-the-art, and show significant improvement on challenging recognition benchmark Pascal VOC 2007. Kernel based fusion is performed to emphasize the complementary nature of our descriptors with respect to the state-of-the-art features.
Keywords :
edge detection; feature extraction; image fusion; image segmentation; object recognition; topology; edge structure; feature pairs; gradient orientation; image edge; kernel based fusion; line based image approximation; line end; object category recognition; segmentation map; topological constraint; Detectors; Feature extraction; Image edge detection; Image segmentation; Kernel; Object recognition; Shape; feature extraction; image representation; object recognition; shape modeling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (ICPR), 2010 20th International Conference on
Conference_Location :
Istanbul
ISSN :
1051-4651
Print_ISBN :
978-1-4244-7542-1
Type :
conf
DOI :
10.1109/ICPR.2010.757
Filename :
5597286
Link To Document :
بازگشت