Output details

A spatio-temporal 2D-models framework for human pose recovery in monocular sequences

<23>The paper describes a novel probabilistic framework for human motion analysis, which takes advantage of 3D information but avoids the use of complex 3D models. It frees 2D dynamical models from having to use human models for practical applications in real environments. The paper led to the authors being invited to give a presentation to a BMVA technical meeting at the BCS. The work was subsequently funded by the Spanish Ministry of Science and Technology through the projects: Human Activity Recognition and Modelling in Real Scenarios (TIN2006-11044) and Human detection, tracking and authentication by facial biometrics and gait analysis (TIC2003-08382-C05-05)