Description
Date depot: 1 janvier 1900
Titre: Machine Identification of Biological Shapes
Directrice de thèse:
Nozha BOUJEMAA (Inria-Paris (ED-130))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
There is currently no existing methodology in image analysis and computer vision which can be applied to highly deformable shape recognition problems of a high complexity such as cells, organs, leaves or flowers.
In fact, the most successful shape recognition algorithms have been developed for rigid or semi-rigid objects, such as faces and cars; moreover, applications involving deformable objects tend to be very specialized, e.g.,identifying handwritten digits or pedestrians.
Consequently, there is a need for generic methodology aimed at category-level recognition of deformable shapes, particularly when both the number of categories and the in-category variation is very large. The goal of this thesis is then to develop such a methodology and corresponding computer algorithms for categorizing deformable shapes.
Doctorant.e: Rejeb Sfar Asma