Holistic and Feature-Based Information Towards Dynamic Multi-Expressions Recognition

May 2010

Holistic and Feature-Based Information Towards Dynamic Multi-Expressions Recognition

Authors:

Zakia Hammal and Corentin Massot

Abstract:

Holistic and feature-based processing have both been shown to be involved differently in the analysis of facial expression by human observer. The current paper proposes a novel method based on the combination of both approaches for the segmentation of "emotional segments" and the dynamic recognition of the corresponding facial expressions. The proposed model is a new advancement of a previously proposed feature-based model for static facial expression recognition (Hammal et al, 2007). First, a new spatial filtering method is introduced for the holistic processing of the face towards the automatic segmentation of "emotional segments". Secondly, the new filtering-based method is applied as a feature-based processing for the automatic and precise segmentation of the transient facial features and estimation of their orientation. Third, a dynamic and progressive fusion process of the permanent and transient facial feature deformations is made inside each "emotional segment" for a temporal recognition of the corresponding facial expression. Experimental results show the robustness of the holistic and feature-based analysis, notably for the analysis of multi-expression sequences. Moreover compared to the static facial expression classification, the obtained performances increase by 12% and compare favorably to human observers' performances.
@conference{Hammal-2010-120275,
author = {Zakia Hammal And Corentin Massot},
title = {Holistic and Feature-Based Information Towards Dynamic Multi-Expressions Recognition},
booktitle = {Proceedings of 5th International Conference on Computer Vision Theory and Applications (VISAPP '10)},
year = {2010},
month = {May},
volume = {2},
pages = {300 - 309},
}
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.