<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>Automatic facial expression recognition from video sequence is an essential research area in the field of computer vision. In this paper, a novel method for recognition facial expressions is proposed, which includes two stages of facial expression feature extraction and facial expression recognition. Firstly, in order to exact robust facial expression features, we use Active Appearance Model (AAM) to extract the global texture feature and optical flow technique to characterize facial expression which is determined facial velocity information. Then, these two features are integrated and converted to visual words using “bag-of-words” models, and facial expression is represented by a number of visual words. Secondly, the Latent Dirichlet Allocation (LDA) model are utilized to classify different facial expressions such as “anger”, “disgust”, “fear”, “happiness”, “neutral”, “sadness”, and “surprise”. The experimental results show that our proposed method not only performs stably and robustly and improves the recognition rate efficiently, but also needs the least dimension when achieves the highest recognition rate , which demonstrates that our proposed method is superior to others.</p></abstract>

Automatic facial expression recognition from video sequence is an essential research area in the field of computer vision. In this paper, a novel method for recognition facial expressions is proposed, which includes two stages of facial expression feature extraction and facial expression recognition. Firstly, in order to exact robust facial expression features, we use Active Appearance Model (AAM) to extract the global texture feature and optical flow technique to characterize facial expression which is determined facial velocity information. Then, these two features are integrated and converted to visual words using “bag-of-words” models, and facial expression is represented by a number of visual words. Secondly, the Latent Dirichlet Allocation (LDA) model are utilized to classify different facial expressions such as “anger”, “disgust”, “fear”, “happiness”, “neutral”, “sadness”, and “surprise”. The experimental results show that our proposed method not only performs stably and robustly and improves the recognition rate efficiently, but also needs the least dimension when achieves the highest recognition rate , which demonstrates that our proposed method is superior to others.

Automatic Recognition of Facial Expression Based on Computer Vision

International Journal on Smart Sensing and Intelligent Systems

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

{"article-title":"Automatic Recognition of Facial Expression Based on Computer Vision"}

Automatic facial expression recognition from video sequence is an essential research area in the field of computer vision. In this paper, a novel method for...

Automatic Recognition of Facial Expression Based on Computer Vision

Data publikacji: 01 wrz 2015

Zakres stron: 1464 - 1483

Otrzymano: 15 kwi 2015

Przyjęty: 01 sie 2015

DOI: https://doi.org/10.21307/ijssis-2017-815

© 2015 Shaoping Zhu et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.