Variational Methods for Normal Integration
QUÉAU, Yvain
Technische Universität Munchen - Technical University Munich - Université Technique de Munich [TUM]
Technische Universität Munchen - Technical University Munich - Université Technique de Munich [TUM]
AUJOL, Jean-François
Institut de Mathématiques de Bordeaux [IMB]
Institut universitaire de France [IUF]
Institut de Mathématiques de Bordeaux [IMB]
Institut universitaire de France [IUF]
QUÉAU, Yvain
Technische Universität Munchen - Technical University Munich - Université Technique de Munich [TUM]
Technische Universität Munchen - Technical University Munich - Université Technique de Munich [TUM]
AUJOL, Jean-François
Institut de Mathématiques de Bordeaux [IMB]
Institut universitaire de France [IUF]
< Reduce
Institut de Mathématiques de Bordeaux [IMB]
Institut universitaire de France [IUF]
Language
en
Article de revue
This item was published in
Journal of Mathematical Imaging and Vision. 2018-05, vol. 60, n° 4, p. 609-632
Springer Verlag
English Abstract
The need for an efficient method of integration of a dense normal field is inspired by several computer vision tasks, such as shape-from-shading, photometric stereo, deflectometry. Inspired by edge-preserving methods from ...Read more >
The need for an efficient method of integration of a dense normal field is inspired by several computer vision tasks, such as shape-from-shading, photometric stereo, deflectometry. Inspired by edge-preserving methods from image processing, we study in this paper several variational approaches for normal integration, with a focus on non-rectangular domains, free boundary and depth discontinuities. We first introduce a new discretization for quadratic integration, which is designed to ensure both fast recovery and the ability to handle non-rectangular domains with a free boundary. Yet, with this solver, discontinuous surfaces can be handled only if the scene is first segmented into pieces without discontinuity. Hence, we then discuss several discontinuity-preserving strategies. Those inspired, respectively, by the Mumford-Shah segmentation method and by anisotropic diffusion, are shown to be the most effective for recovering discontinuities.Read less <
English Keywords
integration
3D-reconstruction
normal field
gradient field
variational methods
photometric stereo
shape-from-shading
Origin
Hal imported