Mixed-state causal modeling for statistical KL-based motion texture tracking

Tomas Crivelli 1, 2 Bruno Cernuschi-Frias 2 Patrick Bouthemy 1 Jian-Feng Yao 1, 3
1 VISTAS - Spatio-Temporal Vision and Learning
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : We are interested in the modeling and tracking of dynamic or motion textures, which refer to dynamic contents that can be classified as a texture with motion (fire, smoke, crowd of people). Experimentally we observe that they depict motion maps with values of a mixed type: a discrete value at zero (absence of motion) and continuous non-null motion values. We thus introduce a temporal mixed-state Markov model for the characterization of motion textures from which a set of 13 parameters is extracted as the descriptive feature of the dynamic content. Then, a motion texture tracking strategy is proposed using the conditional Kullback–Leibler (KL) divergence between mixed-state probability densities, which allows us to estimate the position using a statistical matching approach.
Liste complète des métadonnées

https://hal.inria.fr/inria-00541270
Contributeur : Charles Kervrann <>
Soumis le : mardi 30 novembre 2010 - 11:38:29
Dernière modification le : jeudi 11 janvier 2018 - 06:22:33

Identifiants

Citation

Tomas Crivelli, Bruno Cernuschi-Frias, Patrick Bouthemy, Jian-Feng Yao. Mixed-state causal modeling for statistical KL-based motion texture tracking. Pattern Recognition Letters, Elsevier, 2010, 31 (14), pp.2286-2294. 〈http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V15-50H7D42-1&_user=6068168&_coverDate=10%2F15%2F2010&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1561397025&_rerunOrigin=google&_acct=C000016487&_version〉. 〈10.1016/j.patrec.2010.06.016〉. 〈inria-00541270〉

Partager

Métriques

Consultations de la notice

259