Vector Fields Modelization Using Basis of Polynomials: Application to the Analysis of Simple Face Movements
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) - Mai 2006
In this communication we present an original and general model for the approximation of vector fields and especially displacement vector fields. The proposed method uses the orthonormal multivariate polynomials framework to approximate vector fields as combinations of these particular functions. Then we demonstrate the noise robustness of our model. And finally we show that the model can efficiently be used for the recognition of simple face movement in a webcam acquired sequence.
BibTex references
@InProceedings{DTA2006_2274,
}
author | = {Druon, M. and Tremblais, B. and Augereau, B.}, | |
title | = {Vector Fields Modelization Using Basis of Polynomials: Application to the Analysis of Simple Face Movements.}, | |
booktitle | = {IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)}, | |
month | = {Mai}, | |
year | = {2006}, | |
address | = {Toulouse, France}, | |
keywords | = {vector fields modeling, polynomial basis, movement recognition}, |