RT info:eu-repo/semantics/conferenceObject
T1 Visual object tracking in 3D with color based particle filter
A1 Barrera González, Pablo
A1 Matellán Olivera, Vicente
A1 Cañas, José María
A2 Arquitectura y Tecnologia de Computadores
K1 Informática
K1 Rastreo visual
K1 Método Monte Carlo
K1 Visión múltiple
K1 3D
AB This paper addresses the problem of determining thecurrent 3D location of a moving object and robustly tracking it froma sequence of camera images. The approach presented here uses aparticle  lter and does not perform any explicit triangulation. Onlythe color of the object to be tracked is required, but not any precisemotion model. The observation model we have developed avoidsthe color  ltering of the entire image. That and the MonteCarlotechniques inside the particle  lter provide real time performance.Experiments with two real cameras are presented and lessons learnedare commented. The approach scales easily to more than two camerasand new sensor cues
YR 2012
FD 2012-10-25
LK http://hdl.handle.net/10612/1939
UL http://hdl.handle.net/10612/1939
NO International Conference on Pattern Recognition and Computer Vision, Febrero, 2005, Estambul, Turkia
DS BULERIA. Repositorio Institucional de la Universidad de León
RD 15-jun-2024