Repairing People Trajectories Based on Point Clustering - Computer Science > Computer Vision and Pattern RecognitionReportar como inadecuado




Repairing People Trajectories Based on Point Clustering - Computer Science > Computer Vision and Pattern Recognition - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: This paper presents a method for improving any object tracking algorithmbased on machine learning. During the training phase, important trajectoryfeatures are extracted which are then used to calculate a confidence value oftrajectory. The positions at which objects are usually lost and found areclustered in order to construct the set of -lost zones- and -found zones- inthe scene. Using these zones, we construct a triplet set of zones i.e. threezones: In-Out zone zone where an object can enter or exit the scene -lostzone- and -found zone-. Thanks to these triplets, during the testing phase, wecan repair the erroneous trajectories according to which triplet they are mostlikely to belong to. The advantage of our approach over the existing state ofthe art approaches is that i this method does not depend on a predefinedcontextual scene, ii we exploit the semantic of the scene and iii we haveproposed a method to filter out noisy trajectories based on their confidencevalue.



Autor: Duc Phu Chau INRIA Sophia Antipolis, Francois Bremond INRIA Sophia Antipolis, Etienne Corvee INRIA Sophia Antipolis, Monique Thon

Fuente: https://arxiv.org/







Documentos relacionados