Support Vector Machines for Anatomical Joint Constraint Modelling

Abstract

Abstract-The accurate simulation of anatomical joint models is becoming increasingly important for both realistic animation and diagnostic medical applications. Recent models have exploited unit quaternions to eliminate singularities when modeling orientations between limbs at a joint. This has led to the development of quaternion based joint constraint validation and correction methods. In this paper a novel method for implicitly modeling unit quaternion joint constraints using Support Vector Machines (SVMs) is proposed which attempts to address the limitations of current constraint validation approaches. Initial results show that the resulting SVMs are capable of modeling regular spherical constraints on the rotation of the limb.

Evolved Topology Generalized Multi-layer Perceptron (GMLP) for Anatomical Joint Constraint Modelling

“…Due to inaccuracy in the neural network some correction of valid orientations takes place. Research has shown that an SVM classifier can be used to separate valid orientations from those requiring correction [29]. Recent research [30] has looked employ unsupervised learning more applicable to use with recorded actor or patient specific data (such as that gathered by Herda, Urtasun and Fua [26]).…”