Present at conference from Prof. Carlos Thomaz at INESC Porto.
Title: A simple and efficient supervised method for spatially weighted PCA in Face Image Analysis.
I thought it could be interesting for my project to use in a new lip-sync method, or to give some expression to the puppets. But the conference was very focus in the image reconstruction with PCA components witch wasn’t very useful for me.
Principal Component Analysis (PCA) is an example of a successful unsupervised statistical dimensionality reduction method, which is especially useful in small sample size problems. Despite the well-known attractive properties of PCA, the traditional approach does not incorporate prior information extracted from knowledge of a specific domain. In this talk, I describe a simple and efficient supervised method that allows PCA to incorporate explicitly domain knowledge and generates an embedding space that inherits its optimality properties for dimensionality reduction. The method relies on discriminant weights given by separating hyperplanes to generate the spatially weighted PCA. Several experiments using 2D frontal face images and different data sets have been carried out to illustrate the usefulness of the method for dimensionality reduction, classification and interpretation of face images.
C. E. Thomaz, G. A. Giraldi, J. P. Costa and D. F. Gillies. A Simple and Efficient Supervised Method for Spatially Weighted PCA in Face Image Analysis. Technical Report TR-2010-13, Department of Computing, Imperial College, London, UK, August 2010.