Open PhD Position in Strasbourg (France)

The IGG group, ICube Lab (UMR 7357, University of Strasbourg / CNRS) seeks an excellent student for a 3-years funded PhD position in the field of Computer Graphics starting September 2016.

Candidates are invited to contact us as soon as possible via the two following e-mail addresses: dischler@unistra.fr and allegre@unistra.fr. Candidates must send us the following elements: a detailed CV, marks obtained during Bachelor and Master degree, or Engineering School degree, and a one-page motivation letter. The application deadline is July 8th, 2016.

Please feel free to relay the information.

Title : Extraction and analysis of geometric and photometric properties for the classification of
materials of 3d objects digitized under uncontrolled lighting conditions

Host team : IGG (Computer Graphics and Geometry groupe) at ICube Lab

Advisor : Jean-Michel Dischler, Professor in Computer Science - dischler@unistra.fr

Co-advisor : Rémi Allègre, Associate Professor in Computer Science - allegre@unistra.fr

Prerequisites : Computer graphics and geometric modeling

Abstract : The media content production industry is in increasing demand of tools for creating complex geometric models with realistic appearances more rapidly and more efficiently. While current 3d digitization technologies for real objects considerably simplify the process of creating 3d models, especially for geometry, there is a lot of room for improvement in the field of appearance processing and representation. In this context, it is highly desirable that photometric acquisitions be performed in an uncontrolled lighting environment, and that the appearance of the models be represented by texture layers that fit the requirements of commercially available rendering engines. Existing techniques however only reconstruct approximate reflectance and sophisticated reflectance models do not easily comply with texture layers representations.
The goal of this thesis is to devise a new processing pipeline in two stages. The first stage will consist in accurately extracting the geometric and photometric properties of a 3d object digitized under uncontrolled lighting conditions (geometry, normals, diffuse color, specularity, occlusion), as well as storing them in texture maps. The second stage will consist in analyzing these properties in order to provide a classification of the materials as close as possible to the reality over the whole surface of a digitized object. This classification will be used as input for texture synthesis algorithms developed in the IGG group.

A detailed version of the proposal including bibliography is available at the following address:
https://dpt-info.u-strasbg.fr/~allegre/SUJETS/THESE/SujetTheseIGG2016-JMD-RA_EN.pdf