Recovering people tracking errors using enhanced covariance-based signatures - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

Recovering people tracking errors using enhanced covariance-based signatures

Abstract

This paper presents a new approach for tracking multiple persons in a single camera. This approach focuses on re- covering tracked individuals that have been lost and are detected again, after being miss-detected (e.g. occluded) or after leaving the scene and coming back. In order to correct tracking errors, a multi-cameras re-identification method is adapted, with a real-time constraint. The proposed ap- proach uses a highly discriminative human signature based on covariance matrix, improved using background subtrac- tion, and a people detection confidence. The problem of linking several tracklets belonging to the same individual is also handled as a ranking problem using a learned pa- rameter. The objective is to create clusters of tracklets de- scribing the same individual. The evaluation is performed on PETS2009 dataset showing promising results.
Fichier principal
Vignette du fichier
4797a487.pdf (387.47 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00761322 , version 1 (05-12-2012)

Identifiers

Cite

Julien Badie, Slawomir Bak, Silviu-Tudor Serban, François Bremond. Recovering people tracking errors using enhanced covariance-based signatures. Fourteenth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance - 2012, Jul 2012, Beijing, China. pp.487-493, ⟨10.1109/AVSS.2012.90⟩. ⟨hal-00761322⟩

Collections

INRIA INRIA2
136 View
149 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More