A parallel proximal splitting method for disparity estimation from multicomponent images under illumination variation

Caroline Chaux (1) , Mireille El Gheche (1) , Joumana Farah (2) , Jean-Christophe Pesquet (1) , Beatrice Pesquet-Popescu (3)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 Department of Telecommunications, Faculty of Engineering
3 LTCI - Laboratoire Traitement et Communication de l'Information

Proximal splitting algorithms play a central role in finding the numerical solution of convex optimization problems. This paper addresses the problem of stereo matching of multi-component images by jointly estimating the disparity and the illumination variation. The global formulation being non-convex, the problem is addressed by solving a sequence of convex relaxations. Each convex relaxation is non trivial and involves many constraints aiming at imposing some reg- ularity on the solution. Experiments demonstrate that the method is efficient and provides better results compared with other approaches.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Caroline Chaux, Mireille El Gheche, Joumana Farah, Jean-Christophe Pesquet, Beatrice Pesquet-Popescu. A parallel proximal splitting method for disparity estimation from multicomponent images under illumination variation. Journal of Mathematical Imaging and Vision, 2013, 47 (3), pp.167-178. ⟨10.1007/s10851-012-0361-z⟩. ⟨hal-00733454⟩