NDF: Neural Deformable Fields for Dynamic Human Modelling

Deformable Object Modelling Through Cellular Neural Network

This paper presents a new methodology for the deformable object modelling by drawing an analogy between cellular neural network (CNN) and elastic deformation. The potential energy stored in an elastic body as a result of a deformation caused by an external force is propagated among mass points by the non-linear CNN activity. An improved autonomous CNN model is developed for propagating the ener...

A Cellular Neural Network for Deformable Object Modelling

This paper presents a new methodology for the deformation of soft objects by drawing an analogy between cellular neural network (CNN) and elastic deformation. An improved CNN model is developed to simulate the deformation of soft objects. A finite volume based method is presented to derive the discrete differential operators over irregular nets for obtaining the internal elastic forces. The pro...

Kinematics and Dynamic Modelling of the Human Shoulder

This paper deals with kinematics and dynamics of the human shoulder in order to obtain the workspace related to the reachability of the wrist and applications to determine load for the wheelchair. The model contains six revolute degrees of freedom (D.O.F), two in costoclavicular and three in the shoulder joint and one in the elbow joint, while a six D.O.F model was developed according to given ...

Self-Organizing Dynamic Neural Fields

In this paper, we propose a model of cortical self-organization based on the dynamic field theory. Learning is made through the modification of feed-forward connections using a time invariant learning rule that allows for dynamic (or life-long) learning. This preliminary model suggests that cortical plasticity may be conveyed through feed-forward connections only while cortico-cortical connecti...

Shadow Fields for Deformable Objects