Simultaneous Identity and Expression Recognition SVM+HMM

Two stage classifier approach for identity and facial expression recognition, using low dimensional representation of the geometry of the face.

Face geometry is extracted from input images using Active Appearance Models (AAM - running the Simultaneous Inverse Compositional fitting algorithm) and low dimensional manifolds were then derived using Laplacian EigenMaps (LE) resulting in two types of manifolds, one for model identity and the other for person-specific facial expression.

The first stage uses a multiclass Support Vector Machines (SVM) to establish identity across expression changes.

The second stage deals with person-specific expression recognition, and is composed by a network of seven Hidden Markov Models (HMM) displaced in parallel each one specialized on the several facial emotions analysed.
The decision was made by the sequence that yielded the highest probability.

The left-video shows the input image of a sequence exhibiting a facial emotion and the middle and right videos shows the expression an identity manifolds, respectively. The black line shows the path of the tested emotion over the expression manifold.

See details at http://www.isr.uc.pt/~pedromartins

Category:

Science & Technology

Tags:

• Identity Recognition
• Expression Recognition
• Active Appearance Models (AAM)
• Hidden Markov Models (HMM)