Improving Recognition and Identification of Facial Areas Involved in Non-verbal Communication by Feature Selection
Sheerman-Chase, T, Ong, E-J, Pugeault, N and Bowden, R (2013) Improving Recognition and Identification of Facial Areas Involved in Non-verbal Communication by Feature Selection In: 10th IEEE Conference on Automatic Face and Gesture Recognition (FG), 2013-04-22 - 2013-04-26, Shanghai, China.
Available under License : See the attached licence file.
Meaningful Non-Verbal Communication (NVC) signals can be recognised by facial deformations based on video tracking. However, the geometric features previously used contain a signiﬁcant amount of redundant or irrelevant information. A feature selection method is described for selecting a subset of features that improves performance and allows for the identiﬁcation and visualisation of facial areas involved in NVC. The feature selection is based on a sequential backward elimination of features to ﬁnd a effective subset of components. This results in a signiﬁcant improvement in recognition performance, as well as providing evidence that brow lowering is involved in questioning sentences. The improvement in performance is a step towards a more practical automatic system and the facial areas identiﬁed provide some insight into human behaviour.
|Item Type:||Conference or Workshop Item (Conference Paper)|
|Divisions :||Faculty of Engineering and Physical Sciences > Electronic Engineering|
|Date :||22 April 2013|
|Identification Number :||https://doi.org/10.1109/FG.2013.6553764|
|Depositing User :||Symplectic Elements|
|Date Deposited :||13 Dec 2013 13:46|
|Last Modified :||09 Jun 2014 13:40|
Actions (login required)
Downloads per month over past year