DocumentCode :
2401697
Title :
Action MACH a spatio-temporal Maximum Average Correlation Height filter for action recognition
Author :
Rodriguez, Mikel D. ; Ahmed, Javed ; Shah, Mubarak
Author_Institution :
Comput. Vision Lab., Univ. of Central Florida, Orlando, FL
fYear :
2008
fDate :
23-28 June 2008
Firstpage :
1
Lastpage :
8
Abstract :
In this paper we introduce a template-based method for recognizing human actions called action MACH. Our approach is based on a maximum average correlation height (MACH) filter. A common limitation of template-based methods is their inability to generate a single template using a collection of examples. MACH is capable of capturing intra-class variability by synthesizing a single Action MACH filter for a given action class. We generalize the traditional MACH filter to video (3D spatiotemporal volume), and vector valued data. By analyzing the response of the filter in the frequency domain, we avoid the high computational cost commonly incurred in template-based approaches. Vector valued data is analyzed using the Clifford Fourier transform, a generalization of the Fourier transform intended for both scalar and vector-valued data. Finally, we perform an extensive set of experiments and compare our method with some of the most recent approaches in the field by using publicly available datasets, and two new annotated human action datasets which include actions performed in classic feature films and sports broadcast television.
Keywords :
Fourier transforms; frequency-domain analysis; image recognition; 3D spatiotemporal volume; Clifford Fourier transform; action MACH; action recognition; annotated human action datasets; frequency domain; intra-class variability; spatio-temporal maximum average correlation height filter; template-based method; Computational efficiency; Computer vision; Data analysis; Fourier transforms; Frequency domain analysis; Humans; Image motion analysis; Optical films; Optical filters; Spatiotemporal phenomena;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on
Conference_Location :
Anchorage, AK
ISSN :
1063-6919
Print_ISBN :
978-1-4244-2242-5
Electronic_ISBN :
1063-6919
Type :
conf
DOI :
10.1109/CVPR.2008.4587727
Filename :
4587727
Link To Document :
بازگشت