"Cartoon Face Animation via Machine Learning"

Supervised by Dr ND Lawrence

Abstract

Facial animation is a developing field in computer science and several approaches have been used to achieve this goal. Machine learning it is one of the possible techniques to achieve it. In this paper we are going to use a novel algorithm, the Gaussian Process Latent Variable Model, to achieve the animation of a simple cartoon face. The paper will start from the process of building a dataset of training data acquired manually for our purpose and then it will go through the process of machine learning used on this training data and it will show how the computer created the missing data for the animation. The main efforts have been spent on animating the face, resulting in a very simple interface for the program.