Vous êtes ici : Accueil » Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

23 septembre 2018

Postdoctoral position in Large Scale Tracking in High-Density Crowds


Catégorie : Post-doctorant


Postdoctoral position in Large Scale Tracking in High-Density Crowds

Context and objective

The prospective research work focuses on large scale tracking in high-density crowds, in order to improve the safety during major social events, and to help validate simulation models and evacuation scenarios using real observations.

The researcher will be based at Université Paris Sud, and this work will be performed within the framework of ANR MOHICANS, a research project funded by the French National Research Agency. More details about the context are available here:

http://www.agence-nationale-recherche.fr/Project-ANR-15-CE39-0005

During the project, a large body of work has been devoted up to this point to the problem of pedestrian detection and localization in a high-density context. More details about the datasets and the available detectors which may be used further for tracking are here:

http://hebergement.u-psud.fr/emi/MOHICANS/index.html

Due to the interdisciplinary character of the project, the profile we seek for tackling the tracking aspect is quite open, and we will consider strong applications with a background in computer vision, statistical learning, data fusion or pedestrian traffic modelling. Depending on the candidate's background, the tracking problem may be approached as trajectory learning or as Sequential Monte Carlo based tracking with social and spatial constraints. Tracking may be performed in single or multiple views; for the latter case a good background in robotic vision will be appreciated.

Duration

12 months, which may be further extended to 12 additional months, starting as soon as possible.

How to apply

Send a message with your CV, cover letter and at least one letter of recommendation to:

 

Dans cette rubrique

(c) GdR 720 ISIS - CNRS - 2011-2018.