000149698 001__ 149698
000149698 005__ 20190316234821.0
000149698 0247_ $$2doi$$a10.1109/TPAMI.2009.108
000149698 022__ $$a0162-8828
000149698 02470 $$2ISI$$a000277649100002
000149698 037__ $$aARTICLE
000149698 245__ $$aFrom Canonical Poses to 3D Motion Capture Using a Single Camera
000149698 269__ $$a2010
000149698 260__ $$bInstitute of Electrical and Electronics Engineers$$c2010
000149698 336__ $$aJournal Articles
000149698 520__ $$aWe combine detection and tracking techniques to achieve robust 3D motion recovery of people seen from arbitrary viewpoints by a single and potentially moving camera. We rely on detecting key postures, which can be done reliably, using a motion model to infer 3D poses between consecutive detections, and finally refining them over the whole sequence using a generative model. We demonstrate our approach in the cases of golf motions filmed using a static camera and walking motions acquired using a potentially moving one. We will show that our approach, although monocular, is both metrically accurate because it integrates information over many frames and robust because it can recover from a few misdetections.
000149698 6531_ $$aComputer vision
000149698 6531_ $$aMotion
000149698 6531_ $$aVideo analysis
000149698 6531_ $$a3D scene analysis
000149698 6531_ $$aTracking
000149698 700__ $$aFossati, Andrea
000149698 700__ $$aDimitrijevic, Miodrag
000149698 700__ $$0240235$$aLepetit, Vincent$$g149007
000149698 700__ $$0240252$$aFua, Pascal$$g112366
000149698 773__ $$j32$$k7$$q1165-1181$$tIeee Transactions On Pattern Analysis And Machine Intelligence
000149698 8564_ $$s30703285$$uhttps://infoscience.epfl.ch/record/149698/files/FossatiDLF10.pdf$$yPreprint$$zn/a
000149698 909C0 $$0252087$$pCVLAB$$xU10659
000149698 909CO $$ooai:infoscience.tind.io:149698$$pIC$$particle$$qGLOBAL_SET
000149698 917Z8 $$x169221
000149698 937__ $$aEPFL-ARTICLE-149698
000149698 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000149698 980__ $$aARTICLE