Title :
Probabilistic graphicalmodels for human motion tracking
Author :
Gómez, José I. ; Marín-Jiménez, M.J. ; de la Blanca, N. Pérez
Author_Institution :
Dept. Comput. Sci., Univ. of Jaen, Jaen
Abstract :
Graphical models have proved to be very efficient models for labeling image data. In particular, they have been used to label data samples from human body images. In this paper, a DTG-based graphical model is studied for human-body landmark localization and tracking along the image sequence. Experimental results on human motion databases are shown.
Keywords :
computer graphics; graph theory; image sequences; probability; tracking; visual databases; data sample; decomposable triangulated graph; human motion tracking; image database; image sequence; probabilistic graphical model; Biological system modeling; Computer science; Costs; Graphical models; Humans; Image databases; Image sequences; Labeling; Probability distribution; Tracking;
Conference_Titel :
Image Analysis for Multimedia Interactive Services, 2009. WIAMIS '09. 10th Workshop on
Conference_Location :
London
Print_ISBN :
978-1-4244-3609-5
Electronic_ISBN :
978-1-4244-3610-1
DOI :
10.1109/WIAMIS.2009.5031453
