000163450 001__ 163450
000163450 005__ 20190316235031.0
000163450 037__ $$aCONF
000163450 245__ $$aLearning pattern transformation manifolds with parametric atom selection
000163450 269__ $$a2011
000163450 260__ $$c2011
000163450 336__ $$aConference Papers
000163450 500__ $$aInvited paper
000163450 520__ $$aWe address the problem of building a manifold in order to represent a set of geometrically transformed images by selecting a good common sparse approximation of them with parametric atoms. We propose a greedy method to construct a representative pattern such that the total distance between the transformation manifold of the representative pattern and the input images is minimized. In the progressive construction of the pattern we select atoms from a continuous dictionary by optimizing the atom parameters. Experimental results suggest that the representative pattern built with the proposed method provides an accurate representation of data, where the invariance to geometric transformations is achieved due to the transformation manifold model.
000163450 6531_ $$aManifold learning
000163450 6531_ $$aPattern transformation manifolds
000163450 6531_ $$aDimensionality reduction
000163450 6531_ $$aSparse signal approximations
000163450 6531_ $$aLTS4
000163450 700__ $$0242951$$g185439$$aVural, Elif
000163450 700__ $$aFrossard, Pascal$$0241061$$g101475
000163450 7112_ $$dMay 2-6, 2011$$cSingapore$$aSampling Theory and Applications (SAMPTA)
000163450 8564_ $$uhttps://infoscience.epfl.ch/record/163450/files/ParametricPTM.pdf$$zn/a$$s203417$$yn/a
000163450 909C0 $$xU10851$$0252393$$pLTS4
000163450 909CO $$qGLOBAL_SET$$pconf$$ooai:infoscience.tind.io:163450$$pSTI
000163450 917Z8 $$x185439
000163450 917Z8 $$x101475
000163450 937__ $$aEPFL-CONF-163450
000163450 973__ $$rREVIEWED$$sPUBLISHED$$aEPFL
000163450 980__ $$aCONF