Showing papers on "Facial Action Coding System published in 1995"

Facial expression recognition using a dynamic model and motion energy

Abstract: Previous efforts at facial expression recognition have been based on the Facial Action Coding System (FACS), a representation developed in order to allow human psychologists to code expression from static facial "mugshots." We develop new more accurate representations for facial expression by building a video database of facial expressions and then probabilistically characterizing the facial muscle activation associated with each expression using a detailed physical model of the skin and muscles. This produces a muscle based representation of facial motion, which is then used to recognize facial expressions in two different ways. The first method uses the physics based model directly, by recognizing expressions through comparison of estimated muscle activations. The second method uses the physics based model to generate spatio temporal motion energy templates of the whole face for each different expression. These simple, biologically plausible motion energy "templates" are then used for recognition. Both methods show substantially greater accuracy at expression recognition than has been previously achieved. >

Components and recognition of facial expression in the communication of emotion by actors.

Abstract: This article addresses the issue of the communication of emotion by actors. In Study 1, the facial behavior of 6 actors portraying emotions as felt or unfelt were analyzed with the Facial Action Coding System. Results indicated that the portrayals of felt emotions were closer to the expression of genuine emotion than the portrayals of unfelt emotions for 3 of the 6 emotions under investigation. Study 2 examined the decoding of actors' portrayals from facial behavior. Decoders were found to be very accurate in recognizing the emotional category but not in judging the encoding condition.

Classifying Facial Action

Abstract: The Facial Action Coding System, (FACS), devised by Ekman and Friesen (1978), provides an objective means for measuring the facial muscle contractions involved in a facial expression. In this paper, we approach automated facial expression analysis by detecting and classifying facial actions. We generated a database of over 1100 image sequences of 24 subjects performing over 150 distinct facial actions or action combinations. We compare three different approaches to classifying the facial actions in these images: Holistic spatial analysis based on principal components of graylevel images; explicit measurement of local image features such as wrinkles; and template matching with motion flow fields. On a dataset containing six individual actions and 20 subjects, these methods had 89%, 57%, and 85% performances respectively for generalization to novel subjects. When combined, performance improved to 92%.

Facial Expressions of Patients With Dementia: A Comparison of Two Methods of Interpretation

Abstract: Two methods of interpreting the videotaped facial expressions of four patients with severe dementia of the Alzheimer type were compared. Interpretations of facial expressions performed by means of unstructured naturalistic judgements revealed episodes when the four patients exhibited anger, disgust, happiness, sadness, and surprise. When these episodes were assessed by use of modified version of the Facial Action Coding System, there was, in total, 48% agreement between the two methods. The highest agreement, 98%, occurred for happiness shown by one patient. It was concluded that more emotions could be judged by means of the unstructured naturalistic method, which is based on an awareness of the total situation that facilitates imputing meaning into the patients' cues. It is a difficult task to find a balance between imputing too much meaning into the severely demented patients' sparse and unclear cues and ignoring the possibility that there is some meaning to be interpreted.

Facial responsiveness to odours in normal and pervasively developmentally disordered children.

Abstract: The facial responsiveness of 10 mutic children with pervasive developmental disorder (PDD) and 10 normal (N) children matched for sex and chronological age were covertly videotaped while presented with a set of odours contrasted in hedonic valence. Hedonic ratings of the stimuli were obtained both from the group of N subjects and a panel of adults. Two methods were used to measure facial responses in the same subjects. The first method consisted in an analysis of facial movements with the Facial Action Coding System. Results show that PDD and N subjects displayed distinct action units in response to unpleasant odours. PDD subjects typically displayed muscular actions indexing negative experience, while N subjects showed more smiles. With the second method, odour-elicited facial behaviour was rated by a panel of observers, who were asked to judge whether the subjects were exposed a pleasant, neutral or unpleasant smell. The facial responses to unpleasant odours were classified more accurately in PDD than in N subjects. These findings suggest a functional ability to sense the hedonics attached to odours, but a deficit of socialization of hedonic facial displays in developmentally disordered subjects

Extraction of facial sketch images and expression transformation based on FACS

Abstract: This paper proposes a new method that can be used to extract facial sketch image automatically and to transform facial expression. The extraction of facial features is performed by the following three processing steps: segmentation using color characteristics, detection of feature points of each part of the face, and their approximation using polynomials. They are called feature curves and they constitute a facial sketch image. Expression transformation of the sketch image based on the facial action coding system (FACS) is carried out easily by adjusting the locations of the feature points. The control scheme of the feature point locations is based on the FACS. Experimental results on the extraction of the facial sketch image and their expression transformations are provided.

Modeling human facial expressions at multiple resolutions

Abstract: Human facial expression modeling has been an active research area. Most of the existing systems do not provide an easy way to adjust the model such that different levels of detail of expressions can be modeled. We propose a method for modeling human facial expressions at different resolutions. Our method is based on the facial action coding system (FACS) developed by Ekman and Freisen (1977). Although the facial expressions we are mainly interested in are those related to speech either directly or indirectly, the modeling method can be extended to apply to facial expressions that are not related to speech.

Searching for facial expression by genetic algorithm

Abstract: FACS (Facial Action Coding System) was proposed by the psychologists, Paul Ekman and Wallace V. Friesen, to describe a facial expression in terms of 46 muscular movements, called AUs (Action Units). Here, a mesh model of a human face is defined and each action unit deforms the mesh model according to the corresponding movements specified in the FACS system. The mesh model is deformed to create a facial expression by activating several action units with their intensity values. In this paper, we propose a method using GA (Genetic Algorithm), known widely as a function optimizer, to extract the components of an arbitrary facial expression automatically where the expression may be input by a 3D measuring device or a 2D image. By using these components, we can construct more realistic facial expressions tailored to individual characteristics.