000211260 001__ 211260
000211260 005__ 20190317000249.0
000211260 037__ $$aCONF
000211260 245__ $$aDense image registration and deformable surface reconstruction in presence of occlusions and minimal texture
000211260 269__ $$a2015
000211260 260__ $$c2015
000211260 336__ $$aConference Papers
000211260 520__ $$aDeformable surface tracking from monocular images is well-known to be under-constrained. Occlusions often make the task even more challenging, and can result in failure if the surface is not sufficiently textured. In this work, we explicitly address the problem of 3D reconstruction of poorly textured, occluded surfaces, proposing a framework based on a template-matching approach that scales dense robust features by a relevancy score. Our approach is extensively compared to current methods employing both local feature matching and dense template alignment. We test on standard datasets as well as on a new dataset (that will be made publicly available) of a sparsely textured, occluded surface. Our framework achieves state-of-the-art results for both well and poorly textured, occluded surfaces.
000211260 6531_ $$adense image registration
000211260 6531_ $$adeformable surface reconstruction
000211260 6531_ $$aocclusions
000211260 6531_ $$aminimal texture
000211260 700__ $$0246262$$aNgo, Tien Dat$$g211303
000211260 700__ $$aPark, Sanghyuk
000211260 700__ $$0246577$$aJorstad, Anne Alison$$g225441
000211260 700__ $$0246960$$aCrivellaro, Alberto$$g224226
000211260 700__ $$aYoo, Chang
000211260 700__ $$0240252$$aFua, Pascal$$g112366
000211260 7112_ $$aInternational Conference on Computer Vision (ICCV)$$cSantiago, Chile$$dDecember 13-16, 2015
000211260 8564_ $$s3278406$$uhttps://infoscience.epfl.ch/record/211260/files/dense.pdf$$yn/a$$zn/a
000211260 909C0 $$0252087$$pCVLAB$$xU10659
000211260 909CO $$ooai:infoscience.tind.io:211260$$pconf$$pIC$$qGLOBAL_SET
000211260 917Z8 $$x211303
000211260 917Z8 $$x211303
000211260 917Z8 $$x211303
000211260 917Z8 $$x211303
000211260 917Z8 $$x112366
000211260 937__ $$aEPFL-CONF-211260
000211260 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000211260 980__ $$aCONF