Menu:

 

DYNAMICAL ANIMATION APPLIED TO HUMAN BIOMECHANICS

Researches: Xavier Lligades, Antonio Susin.
Description:
Human Biomechanics is one of the rising topics in medical applications of computer graphics and VR. The increasing possibilities of present hardware and the combination of different data sources like: video cameras, electromyography, dynamometry, force platforms or pressure insoles, etc., can be used to compute significant biomechanics values.    

Using our in house motion capture system, we can study 3D position of optical markers in order to obtain kinematics values: angles, velocities and accelerations. The first problem is to obtain a good correspondence between markers position and the corresponding joint centres. Present attempts to solve this problem are based on approximate the centre position from minimization coefficients computed from statistics values. Therefore, you need to obtain previously a good data base of normality from several known voluntaries. Instead of this approach we prefer the possibility of compute joint centres based on predetermined movements that allow the estimation of these centres with enough precision.

A

Once you have computed the kinematics values, then you can use inverse dynamics in two different steps. From one side, you can use direct dynamic animation using the previous computed values to help the tracking optical process when some markers are occluded due to the present motion.
From the other side, you can compute force and torque acting in each joint. The possibility of quantifying torques acting on one joint due to actual motion, together with some activation values coming from electromyography,  can be used to prevent future injures due to an inappropriate motion or exercise, in a work place scenario or during some athletic training. 


Publications:

Example image - aligned to the right

A. Fernández-Baena , A. Susín, X. Lligadas.Biomechanical Validation of Upper-body and Lower-body Joint Movements of Kinect Motion Capture Data. Procc. International Workshop on Motion Capture and Classification (MC2-2012), 4th International Conference on Intelligent Networking and Collaborative Systems (INCoS-2012), pp. 656-661,(2012).(PDF 2.6Mb)


Example image - aligned to the right

J. E. Ramírez, X. Lligadas, A. Susín Adjusting Animation Rigs to Human-like 3D Models. Proc. AMDO-10, Perales F. Fisher B., Lecture Notes in Computer Science, Vol LNCS 6169, pp 300-310 ,2010


Example image - aligned to the right

J. X. Lligadas, E. Ramírez, A. Susín Velocity Based Controllers for Dynamic Character Animation. Proc. CEDI-10, M Chover, M. Otaduy Ed., Garceta Grupo Editorial, pp 77-86, 2010


References:

[MFC99] F. Multon,, L. France, MP Cani-Gascuel, G. Debunne , Animation of Human Walking: a Survey. http://www-evasion.imag.fr/Publications/1999/MFCD99/

[MM01] M. Meredith S.Maddock, Motion Capture File Formats Explained  
http://www.dcs.shef.ac.uk/intranet/research/resmes/CS0111.pdf

[HFP00] L. Herda, P. Fua, R. Pl¨ankers, R. Boulic and D. Thalmann, Skeleton-Based Motion Capture for Robust Reconstruction of Human Motion. http://vrlab.epfl.ch/Publications/pdf/Herda_and_al_CA_00.pdf

[APF99] N. D'Apuzzoa, R. Plänkersb, P. Fua b, A. Gruena and D. Thalmann, Modeling Human Bodies from Video Sequences.

http://vrlab.epfl.ch/Publications/pdf/Apuzzo_and_al_SPIE_99.pdf

Toni Susin