Bienvenue à TAIMA 2018

Session 2 : Riemannian Computing in Image and Video Analysis

The recent interest to Riemannian approaches in Computer Vision and Pattern Recognition is growing because of their suitable properties to model image and video data either in 2D and 3D domains. Recent examples are modeling skeletal data as trajectories in a Lie group or the Kendall shape space which give rise to representation spaces with non-trivial geometries. Other examples cover the Stiefel and Grassmann manifold for subspace comparison (e.g. in face recognition or human activity modeling) or the cone of SPD matrices to study covariance descriptors or diffusion tensor imaging. The goal of this Special Session is to put more the light on some of these emerging approaches and discuss new challenges in terms Big (Image) data modeling and applications of (Deep) Machine Learning techniques. We are waiting contributions on one of the following topics,

• • Geometric invariants
• • Motion invariants
• • Shape and trajectory analysis
• • Statistical methods in shape analysis
• • SPD manifolds
• • Grassmann manifolds

and related applications in Computer Vision, Biometrics recognition, Medical Image Analysis, Robotics, etc.

Soumission

The Special Session will take place in conjunction with the main conference TAIMA. Submitted papers should be written in english. TAIMA instructions should be carefully followed. Please, send your contributions by e-mail to :
boulbaba DOT benamor AT imt-lille-douai DOT fr
Accepted papers will be published in the proceedings of RFMI’2018 published in the CCIS Series of Springer.

Co-chairs of special session

Prof. Boulbaba Ben Amor : IMT Lille Douai/CRIStAL (UMR CNRS 9189).
Prof. Faouzi Ghorbel : ENSI/CRISTAL pole GRIFT.