Correlation and Interpolation Networks for Real-time Expression Analysis/Synthesis
Trevor J. Darrell, Irfan A. Essa, Alex Pentland · 1994
We describe a framework for real-time tracking of facial expressions that uses neurally-inspired correlation and interpolation methods. A distributed view-based representation is used to characterize facial state, and is computed using a replicated correlation network. The ensemble response of the set of view correlation scores is input to a network based interpolation method, which maps perceptual state to motor control states for a simulated 3-D face model. Activation levels of the motor state correspond to muscle activations in an anatomically derived model. By integrating fast and robust 2-D processing with 3-D models, we obtain a system that is able to quickly track and interpret complex facial motions in real-time. 1 INTRODUCTION An important task for natural and artificial vision systems is the analysis and interpretation of faces. To be useful in interactive systems and in other settings where the information conveyed is of a time critical nature, analysis of facial expression...