Digital Library

cab1

 
Title:      OBTAINING HUMAN BEHAVIOR PATTERNS USING HIDDEN MARKOV MODELS IN VIDEO A VIDEO VIGILANCE CONTEXT
Author(s):      Héctor Gómez, Susana Arias, Ronaldo C. Prati
ISBN:      978-989-8533-06-7
Editors:      Hans Weghorn, Leonardo Azevedo and Pedro Isaías
Year:      2011
Edition:      Single
Keywords:      Video vigilance; human behavior; Hidden Markov Model
Type:      Short Paper
First Page:      481
Last Page:      485
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      This paper uses Hidden Markov Models (HMMs) to obtain patterns of human behavior that can be used to distinguish between a normal and suspicious behavior into a video vigilance application problem. To this end, we have hand labeled the actions (events) that in videos where people are acting in either suspicious or normal manner. These labeled sequences are used to estimate transition probabilities for applying HMMs. These patterns are used to classify new test instances, obtaining classification errors of 17% for normal behavior and 20% for suspicious behavior.
   

Social Media Links

Search

Login