En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)
Portail > Offres > Offre UPR8001-FRELER-003 - Intégration de techniques d'apprentissage et d'optimisation combinatoire pour la résolution de problèmes de vision par ordinateur à grande échelle (H/F)

Integration of machine learning and combinatorial optimization techniques for solving Computer Vision large scale problems

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : lundi 7 novembre 2022

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

General information

Reference : UPR8001-FRELER-003
Workplace : TOULOUSE
Date of publication : Friday, September 16, 2022
Type of Contract : FTC Scientist
Contract Period : 12 months
Expected date of employment : 10 October 2022
Proportion of work : Full time
Remuneration : 2700 to 3100 € brut per month
Desired level of education : PhD
Experience required : 1 to 4 years

Missions

The postdoc recruited on this project will study the existing literature combining Operational Research (OR) and machine learning methods in the field of computer vision, in particular those aiming at applications of detection, localization and reidentification of targets in video streams. In a first step, he/she will pursue the work undertaken in the field of person reidentification, which will allow to consolidate these skills and to quickly valorize first results in conferences of Operational Research and Computer Vision. In a second phase, he/she will address the problem of posture estimation by taking inspiration from the methods already proposed in the literature. Finally, he/she will work on integrating all these techniques in order to propose a complete chain of person detection, re-identification and posture tracking in a multi-camera network. The thesis project will be conducted according to an agile approach aiming at producing more and more successful and realistic context experiments by exploiting the camera network already deployed in the ADREAM building (https://www.laas.fr/public/fr/le-projet-adream) of the LAAS-CNRS and thus propose an advanced proof of concept of the approach, dedicated to the visitors of the laboratory.

Activities

Contributions to 3D vision state-of-the-art techniques
Use of combinatorial optimization solvers
Implementation of algorithms in C/C++/Python

Skills

This offer is specifically dedicated to PhD fellow in Computer Vision with good theoretical background in combinatorial optimization and machine learning.
Other skills:
- Autonomy, teamwork
- Tools for vision and ML: OpenCV, neural network libraries and architectures

Work Context

RAP research team, Robotics department at LAAS-CNRS of Toulouse

Constraints and risks

None

We talk about it on Twitter!