• The KIPS Transactions:PartB
• /
• v.18B no.5
• /
• pp.305-314
• /
• 2011

Research into the visual perception of human emotion to recognize an intention has traditionally focused on emotions of facial expression. Recently researchers have turned to the more challenging field of emotional expressions through body posture or activity. Proposed work approaches recognition of basic emotional categories from body postures using neural model applied visual perception of neurophysiology. In keeping with information processing models of the visual cortex, this work constructs a biologically plausible hierarchy of neural detectors, which can discriminate 6 basic emotional states from static views of associated body postures of activity. The proposed model, which is tolerant to parameter variations, presents its possibility by evaluating against human test subjects on a set of body postures of activities.

• Electronics and Telecommunications Trends
• /
• v.35 no.4
• /
• pp.91-102
• /
• 2020

Recently, generative adversarial networks (GANs) is a field of research that has rapidly emerged wherein many studies conducted shows overwhelming results. Initially, this was at the level of imitating the training dataset. However, the GAN is currently useful in many fields, such as transformation of data categories, restoration of erased parts of images, copying facial expressions of humans, and creation of artworks depicting a dead painter's style. Although many outstanding research achievements have been attracting attention recently, GANs have encountered many challenges. First, they require a large memory facility for research. Second, there are still technical limitations in processing high-resolution images over 4K. Third, many GAN learning methods have a problem of instability in the training stage. However, recent research results show images that are difficult to distinguish whether they are real or fake, even with the naked eye, and the resolution of 4K and above is being developed. With the increase in image quality and resolution, many applications in the field of design and image and video editing are now available, including those that draw a photorealistic image as a simple sketch or easily modify unnecessary parts of an image or a video. In this paper, we discuss how GANs started, including the base architecture and latest technologies of GANs used in high-resolution, high-quality image creation, image and video editing, style translation, content transfer, and technology.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.11a
• /
• pp.111-114
• /
• 2005

Squash & Stretch principle is playing an essential principle for animation action. The application of this principle gives the illusion of weight and volume to an animation character, and makes it possible that an animation action be the smooth and soft by escaping from the stiffness and rigidity. If an action of human or object on animation is expressed like a real world, it seems to be unnatural. Any action without Squash & Stretch will look rigid, uninteresting and not alive. It can be applied to movement of all objects, characters' actions, dialogues and facial expressions with a basic rule of mass, volume and gravity. Any action will not be well expressed without this principle. To be a good animation action, it should be deeply applied in 3D animation, not only 2D animation. Thus, a systemic analysis of Squash & Stretch principle is required.

• Journal of Korea Multimedia Society
• /
• v.18 no.2
• /
• pp.253-259
• /
• 2015

The spread of smartphones enabled the two-way communication through the Internet, and character emoticons added fun and comfort to communication among users. To figure out laughter-arousing factors, a survey about emoticon designs of Kakaotalk Itemstore was conducted, targeting university students in their twenties. An interaction of 'user-preferred emoticons' and 'laughter-arousing emoticons' was analyzed. 90% of 'user-preferred emoticons' were 'humorous emoticons', and 84% of respondents answered that there is a relationship between 'user-preferred emoticons' and 'laughter-arousing emoticons'. Consequently, a high interaction between the 'rank of emoticons' and 'laughter-arousing emoticons' was derived. Also, factors of laughter-arousing emoticons were analyzed by studying the ranks of emoticon from 1st to 8th of Kakaotalk Itemstore. Two-divisional figures reminding pure kids, maximization of emotional expression by omission or exaggeration of mouth, and smoothness by concave curves aroused laughter. Intuitively understandable gestures were employed in terms of action. Furthermore, two-divisional figures' acting with comparably small body, hands and foot to their head, and people-mimicking motions of animals with cuteness and familiarity enabled arousal of laughter. In facial expressions, humorous articulation of sad, busy or mad expression enabled positive communion among users.

• The Journal of the Korea Contents Association
• /
• v.6 no.12
• /
• pp.269-277
• /
• 2006

In this paper, we propose an interactive system that generates emotional expressions from arm gestures. By extracting relevant features from key frames, we can infer emotions from arm gestures. The necessary factor for real-time animation is tremendous frame rates. Thus, we propose processing facial emotion expression with 3D application for minimizing animation time. And we propose a method for matching frames and actions. By matching image sequences of exagerrated arm gestures from participants, they feel that they are communicating directly with the portraits.

• International Journal of Contents
• /
• v.15 no.4
• /
• pp.59-64
• /
• 2019

In recent years, emotion recognition has been an interesting and challenging topic. Compared to facial expressions and speech modality, gesture-based emotion recognition has not received much attention with only a few efforts using traditional hand-crafted methods. These approaches require major computational costs and do not offer many opportunities for improvement as most of the science community is conducting their research based on the deep learning technique. In this paper, we propose an end-to-end deep learning approach for classifying emotions based on bodily gestures. In particular, the informative keyframes are first extracted from raw videos as input for the 3D-CNN deep network. The 3D-CNN exploits the short-term spatiotemporal information of gesture features from selected keyframes, and the convolutional LSTM networks learn the long-term feature from the features results of 3D-CNN. The experimental results on the FABO dataset exceed most of the traditional methods results and achieve state-of-the-art results for the deep learning-based technique for gesture-based emotion recognition.

• Korean Security Journal
• /
• no.2
• /
• pp.211-226
• /
• 1999

Troublesome problems are being caused by hallucinogenic drugs which come not only from illegal narcotics but also from unprohibited general drug-kinds such as cough-remedy agents of textromerupan, namely, Lamina, Rubi-king and the likes. Abuse and misuse of narcotics permeated into and threatens mental health of socially-known peoples, actor, actress, talents and businessmen as well. This study especially dealt with problems related with Mesampetamin, Rurajaepam, Texstrometropan and Karisofrodol which affects mental behaviour, and considered counter-measures through analysis and developed appropriate system of goods circulation for guidance to be complied by the medicine professionalists. The guidance sought here is recommended to the government ministries for further enforcement of crack-down of the drug crimes. Arquesa in method is recommend as a method of identifying behaviors and facial expressions of abusers and misusers of narcotics at the spot in place. This study is directed to curtailing hallucinogenic drugs use for absolute preventive security by detecting out black dealers of narcotics supplying to the abusers and misusers. There is a need of special education program to cut off the flow of narcotics or drugs crimes into official security force as well as private security personnel. This study introduced a priority and emphasis of real augmentation to theory and practice of the modern scientific security which can protect the mental health of the guarded persons as well especially in the light of security safety curriculum at the school.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.19 no.3
• /
• pp.596-599
• /
• 2005

Human health is affected by not only physical conditions but also mental and social well-being. Changes of human emotions show up as gestures, facial expressions and sweating. Human emotions are affected by such automatic nerve system functions as blood pressure, blood circulation speed, heart beats, pupillary reflex, fluid transfusion, muscular contraction and digestive organs, all of which influence the holistic diseases. The Oriental Medicine sees from a perspective of unity of divinity and men that human life activities are united in terms of their physical and mental functions. From such a perspective, human Five Organs are linked with Five Mental(五神) and Seven Emotions(七情), while they are affected by each other, influencing the life activities both directly and indirectly. Based on Confucianism, Sa-Sang Theory argues that human emotions can be categorized into four energy states and therefore, that human diseases and physiological conditions there of may be determined differently depending on the Four Energies(四氣). There seems to be some common points between Sa-Sang Theory and the conventional Oriental Medicine in that human emotions affect individuals' health conditions, so there seems to be much room for mutual complementation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.785-804
• /
• 2017

Due to various factors such as postures, facial expressions and illuminations, face recognition by videos often suffer from poor recognition accuracy and generalization ability, since the within-class scatter might even be higher than the between-class one. Herein we address this problem by proposing a hierarchical cascaded classifier for video face recognition, which is a multi-layer algorithm and accounts for the misclassified samples plus their similar samples. Specifically, it can be decomposed into single classifier construction and multi-layer classifier design stages. In single classifier construction stage, classifier is created by clustering and the number of classes is computed by analyzing distance tree. In multi-layer classifier design stage, the next layer is created for the misclassified samples and similar ones, then cascaded to a hierarchical classifier. The experiments on the database collected by ourselves show that the recognition accuracy of the proposed classifier outperforms the compared recognition algorithms, such as neural network and sparse representation.

• Journal of Families and Better Life
• /
• v.26 no.2
• /
• pp.71-81
• /
• 2008

This study checks if the PITR (Person in the Rain) test can diagnose the stress caused by school-work for middle school students, in that the PITR responses correlates with the degree of stress. The subjects in the study are 483 middle school thirdgraders from 8 middle schools in Daegu city. Research tools used are an Academic Stress Scale and a PITR test. For data analysis is adopted with t-test and Chi-square test. The results from the study are as follows. First, the stress of middle students scores 2.47/4 that means that the school stress is average, and there is no gender difference in scores. Second, PITR test shows that boys draw more rain than girls, straight rain with stronger wind, hide face more, no good facial expressions more often than girls. Girls on the contrary show more stable lines, bigger puddles, place protections more appropriately. Third, those with higher academic stress use more unstable strokes, less number of or insufficient direct protections than those with less academic stress, and draw indirect protection such as umbrellas before rain or clouds. The study concludes that a PITR test can be useful in diagnosing the academic stress of middle school students.