Dataset Information

Dynamics of facial actions for assessing smile genuineness.

ABSTRACT: Applying computer vision techniques to distinguish between spontaneous and posed smiles is an active research topic of affective computing. Although there have been many works published addressing this problem and a couple of excellent benchmark databases created, the existing state-of-the-art approaches do not exploit the action units defined within the Facial Action Coding System that has become a standard in facial expression analysis. In this work, we explore the possibilities of extracting discriminative features directly from the dynamics of facial action units to differentiate between genuine and posed smiles. We report the results of our experimental study which shows that the proposed features offer competitive performance to those based on facial landmark analysis and on textural descriptors extracted from spatial-temporal blocks. We make these features publicly available for the UvA-NEMO and BBC databases, which will allow other researchers to further improve the classification scores, while preserving the interpretation capabilities attributed to the use of facial action units. Moreover, we have developed a new technique for identifying the smile phases, which is robust against the noise and allows for continuous analysis of facial videos.

SUBMITTER: Kawulok M

PROVIDER: S-EPMC7785114 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Dynamics of facial actions for assessing smile genuineness.

Kawulok Michal M Nalepa Jakub J Kawulok Jolanta J Smolka Bogdan B

PloS one 20210105 1

Applying computer vision techniques to distinguish between spontaneous and posed smiles is an active research topic of affective computing. Although there have been many works published addressing this problem and a couple of excellent benchmark databases created, the existing state-of-the-art approaches do not exploit the action units defined within the Facial Action Coding System that has become a standard in facial expression analysis. In this work, we explore the possibilities of extracting ...[more]

PMID: 33400708

Similar Datasets

Project description:The subject's inherent growth pattern can be an effective factor in characteristics of smile. More vertical growth in the posterior maxilla than in the anterior maxilla could result in a changed relationship between the occlusal plane and the curvature of the lower lip upon smile. In order to broaden the understanding of how smile gets affected by growth pattern and the underlying hard tissues, the present study was undertaken to compare smile in various growth patterns, to determine sexual dimorphism, if any; as well as to correlate smile with underlying hard tissues.One hundred and fifty subjects were selected amongst the students in the Dental Institute and from the outpatient department of Department Orthodontics and Dentofacial Orthopedics. Sample selected for the study ranged in the age group of 17 to 25 years. Selected individuals were subjected to lateral head cephalometric radiography in the Department of Oral Medicine and Radiology and videography. Cephalograms were traced and the subjects were divided into horizontal, average, and vertical growth pattern on the basis of GoGn-SN, lower anterior facial height, and Jaraback's ratio. The video clip was downloaded to obtain frame of posed smile. Cephalometric and photographic measurements were recorded and subjected to statistical analysis.The mean values of smile parameters were significantly higher in males as compared to females irrespective of the growth pattern. The mean incisal display, interlabial gap, lower lip to incisal edge distance, upper vertical lip length, and occlusal plane angle was highest in both males and females of vertical facial growth pattern group; whereas, the smile index, posterior corridor (left and right) were less in vertical facial growth pattern group in both males and females. Thus, the parameters in vertical dimension were increased in vertical growers whereas, the parameters in transverse dimension decreased.The facial growth pattern has significant influence on the parameters of smile along with definite sexual dimorphism. The angular and linear parameters, except saddle angle and lower incisor to NB (linear and angular), influenced smile.

Project description:BackgroundTrust is a prerequisite for successful social relations. People tend to form a first impression of people's trustworthiness based on their facial appearance. The sex of the judging person and its congruency with the sex of the judged people influence these appraisals. Moreover, trustworthiness and happiness share some facial features, which has led to studies investigating the interplay between both social judgments. Studies revealed high correlation in judging happiness and trustworthiness across different facial identities. However, studies are missing that investigate whether this relationship exists on a within-subject level and whether in-group biases such as the congruency between the sex of the judging and judged individual influence this relationship. In the present study, we addressed these questions.MethodsData were collected in an online-survey in two separate samples (N = 30, German sample, N = 107 Dutch sample). Subjects assessed the intensity of happiness and trustworthiness expressed in neutral and calm facial expression of the same characters (50% males, 50% females). Statistical analyses comprised rm-Anova designs based on rating scores and estimates of within-subject associations between both judgments.ResultsOur findings replicate high correlations between happiness and trustworthiness ratings across facial identities based on average scores across participants. However, the strength of this association was strongly dependent on the methodological approach and inter-subject variability was high. Our data revealed an in-group advantage for trustworthiness in women. Moreover, the faces' sex and emotional expressions differentially influenced the within-subject correlation between both judgments in men and women.ConclusionOur findings replicate previous studies on the association between happiness and trustworthiness judgments. We extend our understanding of the link between both social judgments by uncovering that within-subject variability is high and influenced by sex and the availability and appraisal of positive emotional facial cues.

Project description:Research into the authenticity of facial emotion expressions often focuses on the physical properties of the face while paying little attention to the role of beliefs in emotion perception. Further, the literature most often investigates how people express a pre-determined emotion rather than what facial emotion expressions people strategically choose to express. To fill these gaps, this paper proposes a non-verbal zero-sum game - the Face X Game - to assess the role of contextual beliefs and strategic displays of facial emotion expression in interpersonal interactions. This new research paradigm was used in a series of three studies, where two participants are asked to play the role of the sender (individual expressing emotional information on his/her face) or the observer (individual interpreting the meaning of that expression). Study 1 examines the outcome of the game with reference to the sex of the pair, where senders won more frequently when the pair was comprised of at least one female. Study 2 examines the strategic display of facial emotion expressions. The outcome of the game was again contingent upon the sex of the pair. Among female pairs, senders won the game more frequently, replicating the pattern of results from study 1. We also demonstrate that senders who strategically express an emotion incongruent with the valence of the event (e.g., smile after seeing a negative event) are able to mislead observers, who tend to hold a congruent belief about the meaning of the emotion expression. If sending an incongruent signal helps to explain why female senders win more frequently, it logically follows that female observers were more prone to hold a congruent, and therefore inaccurate, belief. This prospect implies that while female senders are willing and/or capable of displaying fake smiles, paired-female observers are not taking this into account. Study 3 investigates the role of contextual factors by manipulating female observers' beliefs. When prompted to think in an incongruent manner, these observers significantly improve their performance in the game. These findings emphasize the role that contextual factors play in emotion perception-observers' beliefs do indeed affect their judgments of facial emotion expressions.

Dataset Information

Dynamics of facial actions for assessing smile genuineness.

Publications

Dynamics of facial actions for assessing smile genuineness.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets