CNRS - Centre National de la Recherche Scientifique : UMR8548, Inria Paris-Rocquencourt, DI-ENS - Département d'informatique de l'École normale supérieure
Abstract : Human analysis in images and video is a hard problem due to the large variation in human pose, clothing, camera view-points, lighting and other factors. While the explicit modeling of this variability is difficult, the huge amount of available person images motivates for the implicit, data-driven approach to human analysis. In this work we aim to explore this approach using the large amount of images spanning a subspace of human appearance. We model this subspace by connecting images into a graph and propagating information through such a graph using a discriminatively trained graphical model. We particularly address the problems of human pose estimation and action recognition and demonstrate how image graphs help solving these problems jointly. We report results on still images with human actions from the KTH dataset.
Type de document :
Communication dans un congrès
IEEE International Conference on Image Processing, Sep 2011, Brussels, Belgium. 2011
https://hal.inria.fr/hal-01063329
Contributeur : Suha Kwak
<>
Soumis le : lundi 15 septembre 2014 - 15:13:04
Dernière modification le : vendredi 25 mai 2018 - 12:02:06
Document(s) archivé(s) le : mardi 16 décembre 2014 - 10:21:15
Kumar Raja, Ivan Laptev, Patrick Pérez, Lionel Oisel. Joint Pose Estimation and Action Recognition in Image Graphs. IEEE International Conference on Image Processing, Sep 2011, Brussels, Belgium. 2011. 〈hal-01063329〉
