A Procedure for Automatically Estimating Model Parameters in Optical Motion Capture
Author(s) -
Maurice Ringer,
Joan Lasenby
Publication year - 2002
Publication title -
citeseer x (the pennsylvania state university)
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5244/c.16.73
Subject(s) - skeleton (computer programming) , position (finance) , computer vision , artificial intelligence , computer science , motion capture , motion (physics) , human skeleton , finance , economics , programming language
Model-basedoptical motion capturesystemsrequireknowledgeof the position of the markersrelative to the underlyingskeleton,the lengthsof the skeleton’ s limbs, and which limb eachmarker is attachedto. Thesemodelparametersaretypically assumed andenteredinto thesystemmanually, althoughtechniquesexist for calculatingsomeof them,suchasthepositionof themarkersrelative to theskeleton’ s joints. We presenta fully automaticprocedurefor determiningthesemodelparameters.It tracksthe2Dpositionsof themarkersonthecameras’ imageplanesanddetermineswhich markerslie on eachlimb beforecalculatingthepositionof theunderlyingskeleton.The only assumptionis thattheskeletonconsistsof rigid limbsconnectedwith ball joints. Theproposedsystemis demonstratedonanumberof realdataexamplesandis shown to calculategoodestimatesof themodelparametersin each.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom