This paper presents a methodology to accurately record human finger postures during grasping. The main contribution consists of a kinematic model of the human hand reconstructed via magnetic resonance imaging of one subject that (i) is fully parameterized and can be adapted to different subjects, and (ii) is amenable to in-vivo joint angle recordings via optical tracking of markers attached to the skin. The principal novelty here is the introduction of a soft-tissue artifact compensation mechanism that can be optimally calibrated in a systematic way. The high-quality data gathered are employed to study the properties of hand postural synergies in humans, for the sake of ongoing neuro-science investigations. These data are analyzed and some comparisons with similar studies are reported. After a meaningful mapping strategy has been devised, these data could be employed to define robotic hand postures suitable to attain effective grasps, or could be used as prior knowledge in lower-dimensional, real-time avatar hand animation.
A data-driven kinematic model of the human hand with soft-tissue artifact compensation mechanism for grasp synergy analysis
GABICCINI, MARCO;MARINO, HAMAL;BIANCHI, MATTEO
2013-01-01
Abstract
This paper presents a methodology to accurately record human finger postures during grasping. The main contribution consists of a kinematic model of the human hand reconstructed via magnetic resonance imaging of one subject that (i) is fully parameterized and can be adapted to different subjects, and (ii) is amenable to in-vivo joint angle recordings via optical tracking of markers attached to the skin. The principal novelty here is the introduction of a soft-tissue artifact compensation mechanism that can be optimally calibrated in a systematic way. The high-quality data gathered are employed to study the properties of hand postural synergies in humans, for the sake of ongoing neuro-science investigations. These data are analyzed and some comparisons with similar studies are reported. After a meaningful mapping strategy has been devised, these data could be employed to define robotic hand postures suitable to attain effective grasps, or could be used as prior knowledge in lower-dimensional, real-time avatar hand animation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
