Recognition of human facial expressions using 2-dimensional physical model

This paper presents a new idea for recognizing human facial expressions from an overall pattern of the face, represented in a potential field activated by edges in the input imagery, rather than from changes in the shape of the facial organs or their geometrical relationships. A two dimensional grid, called Potential Net, of which nodes are moved by the image force of the edges and springs connected to their four neighbors is used as a model of the field. Thus, the nodal displacement vectors in the Net represent the overall pattern. Each facial expression is determined as the means of the nodal displacement vectors yielded by images in each training set. Since the dimension of the space spanned by the nodal displacement vectors is too high, it is mapped into a low dimensional space, called Emotion Space, by applying the KL expansion. Unknown expressions in input images are estimated from their mapping into the Emotion Space