Share Email Print
cover

Proceedings Paper

Representing activities with layers of velocity statistics for multiple human action recognition in surveillance applications
Author(s): Fabio Martínez; Antoine Manzanera; Eduardo Romero
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

A novel action recognition strategy in a video-surveillance context is herein presented. The method starts by computing a multiscale dense optical flow, from which spatial apparent movement regions are clustered as Regions of Interest (RoIs). Each ROI is summarized at each time by an orientation histogram. Then, a multilayer structure dynamically stores the orientation histograms associated to any of the found RoI in the scene and a set of cumulated temporal statistics is used to label that RoI using a previously trained support vector machine model. The method is evaluated using classic human action and public surveillance datasets, with two different tasks: (1) classification of short sequences containing individual actions, and (2) Frame-level recognition of human action in long sequences containing simultaneous actions. The accuracy measurements are: 96:7% (sequence rate) for the classification task, and 95:3% (frame rate) for recognition in surveillance scenes.

Paper Details

Date Published: 5 March 2014
PDF: 6 pages
Proc. SPIE 9026, Video Surveillance and Transportation Imaging Applications 2014, 90260G (5 March 2014); doi: 10.1117/12.2042588
Show Author Affiliations
Fabio Martínez, Univ. Nacional de Colombia (Colombia)
Antoine Manzanera, Ecole Nationale Supérieure de Techniques Avancées (France)
Eduardo Romero, Univ. Nacional de Colombia (Colombia)


Published in SPIE Proceedings Vol. 9026:
Video Surveillance and Transportation Imaging Applications 2014
Robert P. Loce; Eli Saber; Ned Lecky, Editor(s)

© SPIE. Terms of Use
Back to Top