Bayesian modelling of sensorimotor systems: Application to handwriting

Estelle Gilet (1, 2) , Julien Diard (3) , Pierre Bessiére (1, 2)

1 LIG - Laboratoire d'Informatique de Grenoble
2 E-MOTION - Geometry and Probability for Motion and Action
3 LPNC - Laboratoire de Psychologie et NeuroCognition

This paper concerns Bayesian modeling of a sensorimotor system. We present a preliminary model of handwriting, in which the representation of letter is abstract, and is a pivot between motor and sensor models. We show how our model allows to solve a variety of tasks, like letter reading, recognizing the writer, and letter writing.

Mots clés: Probabilistic Models and Methods; Movement; Simulation

Domaines: Neurosciences [q-bio.NC]; Neurosciences

Estelle Gilet, Julien Diard, Pierre Bessiére. Bayesian modelling of sensorimotor systems: Application to handwriting. Deuxième conférence française de Neurosciences Computationnelles, "Neurocomp08", Oct 2008, Marseille, France. ⟨hal-00331586⟩