000183530 001__ 183530
000183530 005__ 20190316235548.0
000183530 037__ $$aPOST_TALK
000183530 245__ $$aNon-convex optimization for robust multi-view imaging
000183530 269__ $$a2013
000183530 260__ $$c2013
000183530 336__ $$aPosters
000183530 520__ $$aWe study the multi-view imaging problem where one has to reconstruct a set of l images, representing a single scene, from a few measurements made at different viewpoints. We first express the solution of the problem as the minimizer of a non-convex objective function where one needs to estimate one reference image, l foreground images modeling possible occlusions, and a set of l transformation parameters modeling the inter-correlation between the observations. Then, we propose an alternating descent method that attempts to minimize this objective function and produces a sequence converging to one of its critical points. Finally, experiments show that the method accurately recovers the original images and is robust to occlusions.
000183530 6531_ $$aMulti-view imaging
000183530 6531_ $$aNon-convex optimization
000183530 700__ $$0242927$$g179918$$aPuy, Gilles
000183530 700__ $$aVandergheynst, Pierre$$g120906$$0240428
000183530 7112_ $$dJanuary, 2013$$cVillars-sur-Ollon, Switzerland$$aInternational Biomedical and Astronomical Signal Processing (BASP) Frontiers workshop
000183530 8564_ $$uhttps://infoscience.epfl.ch/record/183530/files/Poster-BASP13-Villars-Robust_joint_reconstruction.pdf$$zn/a$$s3973615$$yn/a
000183530 909C0 $$xU10380$$0252392$$pLTS2
000183530 909CO $$ooai:infoscience.tind.io:183530$$qGLOBAL_SET$$pSTI$$pposter
000183530 917Z8 $$x179918
000183530 917Z8 $$x179918
000183530 917Z8 $$x179918
000183530 937__ $$aEPFL-POSTER-183530
000183530 973__ $$aEPFL
000183530 980__ $$aPOSTER