Journal of Cognitive Neuroscience
An approach to brain thinker type recognition based on facial asymmetry
ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
An expert system for human personality characteristics recognition
ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
Local facial asymmetry for expression classification
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Hi-index | 0.00 |
In this paper, two types of holistic asymmetry representations are introduced: a) asymmetry of facial silhouette AFS, b) asymmetry of facial elements AFE. These asymmetry types in combination with the maintaining aspect ratio facial normalization [3], to the best of our knowledge have not been yet proposed in literature by other authors. To investigate quantitative influence of such approach, classification based on the PCA and the Mahalanobis distance is studied. Performed observations confirm existence of correlation between facial asymmetry and classification rate. The experiment results to demonstrate the proposed approach efficiency and accuracy are provided.