• Categorizing facial expressions: a comparison of computational models

      Shenoy, Aruna; Anthony, Sue; Frank, Ray; Davey, Neil (Springer Link, 2011-09)
      Recognizing expressions is a key part of human social interaction, and processing of facial expression information is largely automatic for humans, but it is a non-trivial task for a computational system. The purpose of this work is to develop computational models capable of differentiating between a range of human facial expressions. Raw face images are examples of high-dimensional data, so here we use two dimensionality reduction techniques: principal component analysis and curvilinear component analysis. We also preprocess the images with a bank of Gabor filters, so that important features in the face images may be identified. Subsequently, the faces are classified using a support vector machine. We show that it is possible to differentiate faces with a prototypical expression from the neutral expression. Moreover, we can achieve this with data that has been massively reduced in size: in the best case the original images are reduced to just 5 components. We also investigate the effect size on face images, a concept which has not been reported previously on faces. This enables us to identify those areas of the face that are involved in the production of a facial expression.
    • A comparison of the performance of humans and computational models in the classification of facial expression

      Shenoy, Aruna; Anthony, Sue; Frank, Ray; Davey, Neil (2009)
      Recognizing expressions are a key part of human social interaction, and processing of facial expression information is largely automatic for humans, but it is a non-trivial task for a computational system. In the first part of the experiment, we develop computational models capable of differentiating between two human facial expressions. We perform pre-processing by Gabor filters and dimensionality reduction using the methods: Principal Component Analysis, and Curvilinear Component Analysis. Subsequently the faces are classified using a Support Vector Machines. We also asked human subjects to classify these images and then we compared the performance of the humans and the computational models. The main result is that for the Gabor pre-processed model, the probability that an individual face was classified in the given class by the computational model is inversely proportional to the reaction time for the human subjects.
    • Discriminating angry, happy and neutral facial expression: a comparison of computational models

      Shenoy, Aruna; Anthony, Sue; Frank, Ray; Davey, Neil (Springer, 2009)
      Recognizing expressions are a key part of human social interaction, and processing of facial expression information is largely automatic for humans, but it is a non-trivial task for a computational system. The purpose of this work is to develop computational models capable of differentiating between a range of human facial expressions. Raw face images are examples of high dimensional data, so here we use two dimensionality reduction techniques: Principal Component Analysis and Curvilinear Component Analysis. We also preprocess the images with a bank of Gabor filters, so that important features in the face images are identified. Subsequently the faces are classified using a Support Vector Machine. We show that it is possible to differentiate faces with a neutral expression from those with a happy expression and neutral expression from those of angry expressions and neutral expression with better accuracy. Moreover we can achieve this with data that has been massively reduced in size: in the best case the original images are reduced to just 5 components with happy faces and 5 components with angry faces.
    • Recognizing facial expressions: a comparison of computational approaches

      Shenoy, Aruna; Gale, Tim M.; Davey, Neil; Christiansen, Bruce; Frank, Ray (Springer, 2008)
      Recognizing facial expressions are a key part of human social interaction,and processing of facial expression information is largely automatic, but it is a non-trivial task for a computational system. The purpose of this work is to develop computational models capable of differentiating between a range of human facial expressions. Raw face images are examples of high dimensional data, so here we use some dimensionality reduction techniques: Linear Discriminant Analysis, Principal Component Analysis and Curvilinear Component Analysis. We also preprocess the images with a bank of Gabor filters, so that important features in the face images are identified. Subsequently the faces are classified using a Support Vector Machine. We show that it is possible to differentiate faces with a neutral expression from those with a smiling expression with high accuracy. Moreover we can achieve this with data that has been massively reduced in size: in the best case the original images are reduced to just 11 dimensions.
    • Representation and classification of facial expression in a modular computational model

      Shenoy, Aruna; Gale, Tim M.; Frank, Ray; Davey, Neil (World Scientific Pub Co Inc, 2008)