000134227 001__ 134227
000134227 005__ 20190316234516.0
000134227 0247_ $$2doi$$a10.1109/ICDSP.2009.5201223
000134227 02470 $$2ISI$$a000276494500006
000134227 037__ $$aCONF
000134227 245__ $$aA Sparsity Constrained Inverse Problem to Locate People in a Network of Cameras
000134227 269__ $$a2009
000134227 260__ $$aAegean island of Santorini, Greece$$c2009
000134227 336__ $$aConference Papers
000134227 520__ $$aA novel approach is presented to locate dense crowd of people in a network of fixed cameras given the severely degraded background subtracted silhouettes. The problem is formulated as a sparsity constrained inverse problem using an adaptive dictionary constructed on- line. The framework has no constraint on the number of cameras neither on the surface to be monitored. Even with a single camera, partially occluded and grouped people are correctly detected and segmented. Qualitative results are presented in indoor and outdoor scenes.
000134227 6531_ $$aInverse problem
000134227 6531_ $$aSparsity
000134227 6531_ $$aPeople detection
000134227 6531_ $$alts2
000134227 6531_ $$alts4
000134227 700__ $$0242925$$aAlahi, Alexandre$$g129343
000134227 700__ $$aBoursier, Yannick
000134227 700__ $$0243987$$aJacques, Laurent$$g182131
000134227 700__ $$0240428$$aVandergheynst, Pierre$$g120906
000134227 7112_ $$a16th International Conference on Digital Signal Processing$$cAegean island of Santorini, Greece$$dJuly 5-7, 2009
000134227 773__ $$t16th International Conference on Digital Signal Processing
000134227 8564_ $$uhttp://www.dsp2009.org$$zURL
000134227 8564_ $$s7022$$uhttps://infoscience.epfl.ch/record/134227/files/atoms.jpg$$yn/a$$zn/a
000134227 8564_ $$s1395499$$uhttps://infoscience.epfl.ch/record/134227/files/dsp09.pdf$$yn/a$$zn/a
000134227 909C0 $$0252393$$pLTS4$$xU10851
000134227 909C0 $$0252392$$pLTS2$$xU10380
000134227 909C0 $$0252606$$pVITA$$xU13529
000134227 909CO $$ooai:infoscience.tind.io:134227$$pconf$$pSTI$$pENAC$$qGLOBAL_SET
000134227 917Z8 $$x173008
000134227 937__ $$aEPFL-CONF-134227
000134227 973__ $$aEPFL$$rNON-REVIEWED$$sPUBLISHED
000134227 980__ $$aCONF