DocumentCode :
594991
Title :
Learning statistically relevant edge structure improves low-level visual descriptors
Author :
Tabernik, D. ; Kristan, Matej ; Boben, Marko ; Leonardis, Ale
fYear :
2012
fDate :
11-15 Nov. 2012
Firstpage :
1471
Lastpage :
1474
Abstract :
Over the recent years, low-level visual descriptors, among which the most popular is the histogram of oriented gradients (HOG), have shown excellent performance in object detection and categorization. We form a hypothesis that the low-level image descriptors can be improved by learning the statistically relevant edge structures from natural images. We validate this hypothesis by introducing a new descriptor called the histogram of compositions (HoC). HoC exploits a learnt vocabulary of parts from a state-of-the-art hierarchical compositional model. Furthermore, we show that HoC is a complementary HoC descriptor to HOG. We experimentally compare our descriptor to the popular HOG descriptor on the task of object categorization. We have observed approximately 4% improved categorization performance of HoC over HOG at lower dimensionality of the descriptor. Furthermore, in comparison to HOG, we show a categorization improvement of approximately 10% when combining HOG with the proposed HoC.
Keywords :
learning (artificial intelligence); object detection; statistical analysis; HOG; HoC; hierarchical compositional model; histogram of compositions; histogram of oriented gradients; low-level image descriptors; low-level visual descriptors; natural images; object categorization; object detection; statistically relevant edge structure learning; Accuracy; Histograms; Image edge detection; Kernel; Libraries; Object detection; Shape;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (ICPR), 2012 21st International Conference on
Conference_Location :
Tsukuba
ISSN :
1051-4651
Print_ISBN :
978-1-4673-2216-4
Type :
conf
Filename :
6460420
Link To Document :
بازگشت