A Fully Automatic System To Model Faces From a Single Image
Zicheng Liu · 2003
Abstract We present a system which automatically generates a 3D face model from a single frontal image of a face. Our system consists of two components. The rst component is the feature feature detection, and the second component is the model tting. We use an existing feature detection software for the rst component. After we detect the face features, we t a 3D face model by using a linear space of face geometries and assuming an orthogonal projection. Even though the depths of the resulting face models are usually not accurate, the models look recognizable as long as the view angle is not too far from the front. Our system has the advantage that it is fully automatic, robust, and fast. It can be used in a variety of applications for which the accuracy of depths are not critical such as games, avatars, face recognition on close-to-front-view images, etc. 1