A Tensor Based on Optical Flow for Global Description of Motion in Videos

Author

Mota, Virginia F. ; Perez, E.A. ; Vieira, Marcelo B. ; Maciel, Luiz M. ; Precioso, Frederic ; Gosselin, P.H.

Author_Institution

DCC/ICE, Univ. Fed. de Juiz de Fora, Juiz de Fora, Brazil

fYear

2012

fDate

22-25 Aug. 2012

Firstpage

298

Lastpage

301

Abstract

Motion is one of the main characteristics that describe the semantic information of videos. In this work, a global video descriptor based on orientation tensors is proposed. This descriptor is obtained by combining polynomial coefficients calculated for each image in a video. The coefficients are found through the projection of the optical flow on Legendre polynomials, reducing the dimension of per frame motion estimations. The sequence of coefficients are then combined using orientation tensors. The global tensor descriptor created is evaluated by a classification of the KTH video database with a SVM classifier.

Keywords

Legendre polynomials; image classification; image sequences; motion estimation; support vector machines; tensors; video signal processing; KTH video database classification; Legendre polynomials; SVM classifier; dimension reduction; global motion description; global tensor descriptor; global video descriptor; optical flow; orientation tensors; per frame motion estimations; polynomial coefficients; semantic video information; Approximation methods; Humans; Optical imaging; Polynomials; Tensile stress; Vectors; Videos; Global descriptor; Motion modeling; Optical Flow; Orientation tensor; SVM;

fLanguage

English

Publisher

ieee

Conference_Titel

Graphics, Patterns and Images (SIBGRAPI), 2012 25th SIBGRAPI Conference on

Conference_Location

Ouro Preto

ISSN

1530-1834

Print_ISBN

978-1-4673-2802-9

Type

conf

DOI

10.1109/SIBGRAPI.2012.48

Filename

6382770