Human action recognition from local part model

Author

Shi, Feng ; Petriu, Emil M. ; Cordeiro, Albino

Author_Institution

Sch. of Electr. Eng. & Comput. Sci. (EECS), Univ. of Ottawa, Ottawa, ON, Canada

fYear

2011

fDate

14-17 Oct. 2011

Firstpage

35

Lastpage

38

Abstract

In this paper, we present a part model for human action recognition from video. We use 3D HOG descriptor and bag-of-feature to represent video. To overcome the unordered events of bag-of-feature approach, we propose a novel multiscale local part model to preserve temporal context. Our method builds upon several recent ideas including dense sampling, local spatial-temporal (ST) features, 3D HOG descriptor, BOF representation and non-linear SVMs. The preliminary results on KTH action dataset show a higher recognition rate than recent studies.

Keywords

feature extraction; image recognition; image representation; image sampling; image sequences; support vector machines; video signal processing; 3D HOG descriptor; BOF representation; KTH action dataset; bag-of-feature; computer vision; dense sampling; human action recognition; image sequences; local spatial-temporal features; multiscale local part model; nonlinear SVM; video representation; Computational modeling; Feature extraction; Histograms; Humans; Support vector machines; Three dimensional displays; Visualization; 3D HOG descriptor; action recognition; bag-of-feature (BOF); local spatio-temporal (ST) features; part model;

fLanguage

English

Publisher

ieee

Conference_Titel

Haptic Audio Visual Environments and Games (HAVE), 2011 IEEE International Workshop on

Conference_Location

Hebei

Print_ISBN

978-1-4577-0500-7

Type

conf

DOI

10.1109/HAVE.2011.6088408

Filename

6088408