Vous êtes ici : Accueil » Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

4 juin 2018

PhD position in Image Processing and Machine Learning


Catégorie : Doctorant


LuxScan Technologies is a world-wide leader in the design, development and installation of industrial scanners for automation in the timber industry. Our company, based in Luxemburg, is part of the WEINIG group. WEINIG is a German company, producing machines and systems for solid wood and panels production.

LITIS is a computer science laboratory of Normandie Université in Rouen, France. It is well-known center in the research fields of Machine Learning and Computer Vision with more than 110 researchers and 60 PhD students.

LuxScan and LITIS are looking for talented, skilled and motivated student, interested by a PhD in the field of Image processing and machine learning The research will be focused on Deep learning, applied on Wood material.

 

Responsibilities

The goal of the PhD work will be to perform research in the field of deep learning applied to the classification and segmentation of wood material singularities with real time performance constraints.

Architecture of the system may involve (not limited to) Fully Convolutional Networks (FCN), Recurrent Neural Networks (RNN) including LSTM for classification and segmentation; Generative Adversarial Network (GAN) for data augmentation. Data consists in multi-modal and multidimensional pictures, for example acquired with linear camera. One of the challenge of this work is to take into account the distribution shift between timber types, as some species may not be available during training phase.

Education and experience

Benefits

We offer :

Starting date

September / October 2018

Job location

The job is based in Luxemburg. Thanks to send your resume to following address:

LuxScan Technologies
Rue de l’industrie
L-3895 Foetz
Luxembourg

Or by email: jobs@luxscan.com with the reference : LUXRND-PhD-2018

 

Dans cette rubrique

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