Regression-Based Multi-view Facial Expression Recognition

We present a regression-based scheme for multi-view facial expression recognition based on 2-D geometric features. We address the problem by mapping facial points (e.g. mouth corners) from non-frontal to frontal view where further recognition of the expressions can be performed using a state-of-the-art facial expression recognition method. To learn the mapping functions we investigate four regression models: Linear Regression (LR), Support Vector Regression (SVR), Relevance Vector Regression (RVR) and Gaussian Process Regression (GPR). Our extensive experiments on the CMU MultiPIE facial expression database show that the proposed scheme outperforms view-specific classifiers by utilizing considerably less training data.

• Type: Journal
• Year: 2010
• Where: ICPR
• Authors: Ognjen Rudovic, Ioannis Patras, Maja Pantic
• Keywords: Computer Vision, Facial Expression Recognition, ICPR 2010, Multi-view Facial Expression, Vector Regression