Space-time spectral model for object detection in dynamic textured background

被引:14
作者
Ali, Imtiaz [1 ,2 ,4 ]
Mille, Julien [2 ,3 ]
Tougne, Laure [1 ,2 ]
机构
[1] Univ Lyon 2, LIRIS, UMR5205, F-69676 Lyon, France
[2] Univ Lyon, CNRS, Lyon, France
[3] Univ Lyon 1, LIRIS, UMR5205, F-69622 Villeurbanne, France
[4] Opt Labs, Islamabad, Pakistan
关键词
Background model; Local Fourier transform; Dynamic texture; Object detection; SEGMENTATION; CLASSIFICATION; SCALE;
D O I
10.1016/j.patrec.2012.06.011
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Background models are used for object detection in many computer vision algorithms. In this article, we propose a novel background modeling method based on frequency for spatially varying and time repetitive textured background. The local Fourier transform is applied to construct a pixel-wise representation of local frequency components. We apply our method for object detection in moving background conditions. Experimental results of our frequency-based background model are evaluated both qualitatively and quantitatively. (c) 2012 Elsevier B.V. All rights reserved.
引用
收藏
页码:1710 / 1716
页数:7
相关论文
共 25 条
[1]  
Abraham B., 2005, Proc. of the International Workshop on Texture Analysis and Synthesis, P53
[2]  
[Anonymous], IEEE COMPUT VISION P
[3]  
[Anonymous], P CAIP
[4]   MULTICHANNEL TEXTURE ANALYSIS USING LOCALIZED SPATIAL FILTERS [J].
BOVIK, AC ;
CLARK, M ;
GEISLER, WS .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1990, 12 (01) :55-73
[5]   Modeling, clustering, and segmenting video with mixtures of dynamic textures [J].
Chan, Antoni B. ;
Vasconcelos, Nuno .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2008, 30 (05) :909-926
[6]   Clustering Dynamic Textures with the Hierarchical EM Algorithm [J].
Chan, Antoni B. ;
Coviello, Emanuele ;
Lanckriet, Gert. R. G. .
2010 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2010, :2022-2029
[7]   Dynamic textures [J].
Doretto, G ;
Chiuso, A ;
Wu, YN ;
Soatto, S .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2003, 51 (02) :91-109
[8]   Dynamic shape and appearance models [J].
Doretto, Gianfranco ;
Soatto, Stefano .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2006, 28 (12) :2006-2019
[9]   Background and foreground modeling using nonparametric kernel density estimation for visual surveillance [J].
Elgammal, A ;
Duraiswami, R ;
Harwood, D ;
Davis, LS .
PROCEEDINGS OF THE IEEE, 2002, 90 (07) :1151-1163
[10]  
Ghanem B., 2007, IEEE INTERNAT C COMP, P1