• Journal of Information Science Theory and Practice
• /
• v.7 no.2
• /
• pp.32-39
• /
• 2019

A convolutional neural network (CNN) has been widely used in facial expression recognition (FER) because it can automatically learn discriminative appearance features from an expression image. To make full use of its discriminating capability, this paper suggests a simple but effective method for CNN based FER. Specifically, instead of an original expression image that contains facial appearance only, the expression image with facial geometry visualization is used as input to CNN. In this way, geometric and appearance features could be simultaneously learned, making CNN more discriminative for FER. A simple CNN extension is also presented in this paper, aiming to utilize geometric expression change derived from an expression image sequence. Experimental results on two public datasets (CK+ and MMI) show that CNN using facial geometry visualization clearly outperforms the conventional CNN using facial appearance only.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.6
• /
• pp.1137-1144
• /
• 2022

In this study, an artificial intelligence(AI) was developed to help with facial expression practice in order to express emotions. The developed AI used multimodal inputs consisting of sentences and facial images for deep neural networks (DNNs). The DNNs calculated similarities between the emotions predicted by the sentences and the emotions predicted by facial images. The user practiced facial expressions based on the situation given by sentences, and the AI provided the user with numerical feedback based on the similarity between the emotion predicted by sentence and the emotion predicted by facial expression. ResNet34 structure was trained on FER2013 public data to predict emotions from facial images. To predict emotions in sentences, KoBERT model was trained in transfer learning manner using the conversational speech dataset for emotion classification opened to the public by AIHub. The DNN that predicts emotions from the facial images demonstrated 65% accuracy, which is comparable to human emotional classification ability. The DNN that predicts emotions from the sentences achieved 90% accuracy. The performance of the developed AI was evaluated through experiments with changing facial expressions in which an ordinary person was participated.

• Journal of dental hygiene science
• /
• v.7 no.2
• /
• pp.89-96
• /
• 2007

The purpose of this study was to examine the relationship between the self-expression level of dental hygiene students related to communicative competence and their stress during clinical practice and what affected their stress. The subjects in this study were 125 dental hygiene students in W college, on whom a survey was conducted from September 18 through 30, 2006. After the collected data were analyzed with SPSS WIN 10.0 program, the following findings were acquired: 1. When a factor analysis was made to evaluate the self-expression of the students, there appeared three different categories of self-expression: voice/content, facial expression/attitude and sentiment. The three made a 58.1% prediction of their self-expression. As for overall reliability, they turned out highly reliable(Cronbach'a = .881). 2. The dental hygiene students got a mean of 3.58 out of possible five points in self-expression, which indicated that they expressed themselves relatively well. Concerning connections between their general characteristics and self-expression level, those who were inactive during clinical practice got a mean of 3.28, whereas the others who were active got a mean of 3.85. It implied that those who took a more active attitude to clinical practice expressed themselves better(p < .01). The person with whom they found it hard to get along made a statistically significant difference to their self-expression(p < .05). The students who didn't fare well with dental hygienists got the best score(3.70). The second best group(3.53) didn't get along with dentists, followed by assistant nurses(3.46) and patients/caregivers(3.31). As for the impact of the field of dream job, the students who hoped to work or study overseas(4.21) excelled in self-expression those who wanted to be hired in a general hospital, to go onto a school of higher grade and to work in a public dental clinic(p < .05). Among the general characteristics, satisfaction level with major, health status and motivation of choosing dental hygiene made no statistically significant differences to their self-expression. 3. Regarding relations between self-expression level and stress about clinical practice, those who didn't express themselves properly in terms of sentiment scored higher in stress level(3.65). Their stress was statistically significantly different according to self-expression level (p < .05). 4. As for the influence of self-expression and general characteristics on stress with clinical practice, sentiment was selected from among the self-expression categories as a decisive factor to affect stress. Their stress varied statistically significantly with that(p < .05). In contrast, their demographic variables made no statistically significant difference to that, which made a 79.2% prediction of it.

• (The) Research of the performance art and culture
• /
• no.18
• /
• pp.223-255
• /
• 2009

This paper deals with the tradition and identity of physical expression(${\bar{a}}{\dot{n}}gika$ abhinaya) system in Indian traditional theatre and dance. The ${\bar{a}}{\dot{n}}gika$ abhinaya of Indian dance and theatre has been stylized through long terms and thereby defines their identity and peculiarity by its highly stylized expression techniques while following dramatic way($n{\bar{a}}tya$-dharmi). The ${\bar{a}}{\dot{n}}gika$ abhinaya is said to had been derived from the ritualistic practice of ancient Vedic period(B.C. 15th-B.C. 5th c.), which is most clearly exemplified in hand gesture(hasta mudra). In $N{\bar{a}}tya$ ${\acute{S}}astra$, ${\bar{a}}{\dot{n}}gika$ abhinaya is explained in detail. It is classified broadly into facial expression(mukhaja abhinaya), gesture expression(śārīra abhinaya), movement expression($cest{\bar{a}}krita$ abhinaya). Further, $N{\bar{a}}tya$ ${\acute{S}}astra$ divides their various usages by parts which include head, glance, eyes, nose, eyelid, cheek, lower lip, chin, mouth, hand, chest, waist, belly, hip. thigh, claf, foot. Besides, it explains diverse ways of foot movement($c{\bar{a}}ri$), standing poses(sthana), gait(gati) and their combined movements(mandala). Many forms of Indian traditional theatre and dance basically follows the rule of ${\bar{a}}{\dot{n}}gika$ abhinaya proposed in $N{\bar{a}}tya$ ${\acute{S}}astra$, which identifies their unique characteristics.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.992-998
• /
• 2018

Presentation is a technique that is logically and confidently conveying your thoughts and opinions in front of others. It is essential when you go to school or work. However, it takes a lot of time, money, and effort to improve your presentation skills. In this paper, we describe the development of a presentation coaching application that analyzes the presentation practice video. The application program can analyze the presentation time, the speaker's expression, the use of duplicate words, etc.

• Archives of Craniofacial Surgery
• /
• v.22 no.1
• /
• pp.1-10
• /
• 2021

Botulinum toxin type A (BoNT-A), onabotulinumtoxinA (Botox) was approved by the United States Food and Drug Administration for temporary improvement of glabellar lines in patients 65 years and younger in 2002, and has also been used widely for aesthetic purposes such as hyperhidrosis, body shape contouring, and other noninvasive facial procedures. BoNT-A inhibits presynaptic exocytosis of acetylcholine (ACh)-containing vesicles into the neuromuscular junction at cholinergic nerve endings of the peripheral nervous system, thereby paralyzing skeletal muscles. ACh is the most broadly used neurotransmitter in the somatic nervous system, preganglionic and postganglionic fibers of parasympathetic nerves, and preganglionic fibers or postganglionic sudomotor nerves of sympathetic nerves. The scientific basis for using BoNT-A in various cosmetic procedures is that its function goes beyond the dual role of muscle paralysis and neuromodulation by inhibiting the secretion of ACh. Although the major target organs for aesthetic procedures are facial expression muscles, skeletal body muscles, salivary glands, and sweat glands, which are innervated by the somatic or autonomic nerves of the peripheral cholinergic nerve system, few studies have attempted to directly explain the anatomy of the areas targeted for injection by addressing the neural physiology and rationale for specific aesthetic applications of BoNT-A therapy. In this article, we classify the various cosmetic uses of BoNT-A according to the relevant component of the peripheral nervous system, and describe scientific theories regarding the anatomy and physiology of the cholinergic nervous system. We also review critical physiological factors and conditions influencing the efficacy of BoNT-A for the rational aesthetic use of BoNT-A. We hope that this comprehensive review helps promote management policies to support long-term, safe, successful practice. Furthermore, based on this, we look forward to developing and expanding new advanced indications for the aesthetic use of BoNT-A in the future.