Recognizing upper face action units for facial expression analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001

Proceedings of the 7th International Conference on Methods and Techniques in Behavioral Research - MB '10, 2010

2000

Most of the current work on automated facial expression analysis attempt to recognize a small set of prototypic expressions, such as joy and fear. Such prototypic expressions, however, occur infrequently, and human emotions and intentions are communicated more often by changes in one or two discrete features. To capture the full range of facial expression, detection, tracking, and classification of fine-grained changes in facial features are needed.

2006

2018

Face expressions recognition has recently received a significant amount of attention. Over the past 30 years a lot of paper were written describing different features used in face expressions recognition system. This paper provides an overview of all these features and describes some of them in detail. Action Units (AU) are defined and summarized as part of the Facial Action Coding System (FACS). They are significant components regarding face description. There are three underlying motivations for me to write this paper. First of all, I want to give a short review of how the subject of face expressions recognition came to be and where this topic it is headed. The second motivation is to reveal the differences between various approaches used in each step of the face expressions recognition system and finally to show the challenges and opportunities in this field.

Many approaches to facial expression recognition focus on assessing the six basic emotions (anger, disgust, happiness, fear, sadness, and surprise). Real-life situations proved to produce many more subtle facial expressions. A reliable way of analyzing the facial behavior is the Facial Action Coding System (FACS) developed by Ekman and Friesen, which decomposes the face into 46 action units (AU) and is usually performed by a human observer. Each AU is related to the contraction of one or more specific facial muscles. In this study we present an approach towards automatic AU recognition enabling recognition of an extensive palette of facial expressions. As distinctive features we used motion flow estimators between every two consecutive frames, calculated in special regions of interest (ROI). Even though a lot has been published on the facial expression recognition theme, it is still difficult to draw a conclusion regarding the best methodology as there is no common basis for compari...

International Journal of Engineering & Technology

Most works in quantifying facial deformation are based on action units (AUs) provided by the Facial Action Coding System (FACS) which describes facial expressions in terms of forty-six component movements. AU corresponds to the movements of individual facial muscles. This paper presents a rule based approach to classify the AU which depends on certain facial features. This work only covers deformation of facial features based on posed Happy and the Sad expression obtained from the BU-4DFE database. Different studies refer to different combination of AUs that form Happy and Sad expression. According to the FACS rules lined in this work, an AU has more than one facial property that need to be observed. The intensity comparison and analysis on the AUs involved in Sad and Happy expression are presented. Additionally, dynamic analysis for AUs is studied to determine the temporal segment of expressions, i.e. duration of onset, apex and offset time. Our findings show that AU15, for sad exp...

2004

2009

Abstract. This paper concentrates on the comparisons of systems that are used for the recognition of expressions generated by six upper face action units (AUs) by using Facial Action Coding System (FACS). Haar wavelet, Haar-Like and Gabor wavelet coefficients are compared, using Adaboost for feature selection. The binary classification results by using Support Vector Machines (SVM) for the upper face AUs have been observed to be better than the current results in the literature, for example 96.5% for AU2 and 97.6% for AU5.

… of California at San …, 2001