• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 1999.11a
• /
• pp.97-100
• /
• 1999

Human feels kinesthetic sense in response to the force acted on him. In order to represent kinesthetic sense, a force is analyzed as mechanical impedance (i.e., stiffness or damping) and implemented by active impedance control. In this research, a 3-dimensional arm motion generator is developed to present various mechanical impedance characteristics to an operator. An introduction of virtual reality provides not only a visual effect in virtual environment but also the change in force synchronized with the visual effect in real time.

• Science of Emotion and Sensibility
• /
• v.11 no.4
• /
• pp.637-644
• /
• 2008

Human emotion has been tried to be recognized by physiological measurements in developing emotion machine enabling to understand and react to user's emotion. This study is to find the time-dependent physiological measurements and their variation characteristics for discriminating emotions according to dimensional emotion model. Ten university students were asked to watch sixteen prepared images to evoke different emotions. Their subjective emotions and autonomic nervous responses such as ECG (electrocardiogram), PPG (photoplethysmogram), GSR (Galvanic skin response), RSP (respiration), and SKT(skin temperature) were measured during experiment. And these responses were analyzed into HR(Heart Rate), Respiration Rate, GSR amplitude average, SKT amplitude average, PPG amplitude, and PTT(Pulse Transition Time). TDPs(Time dependent parameters) defined as the delay, the activation, the half recovery and the full recovery of respective physiological signal in this study have been determined and statistically compared between variations from different emotions. The significant tendencies in TDP were shown between emotions. Therefore, TDP may provide useful measurements with emotion recognition.

• Journal of Korea Game Society
• /
• v.17 no.3
• /
• pp.45-52
• /
• 2017

This paper proposes a new strategy of emotion-based music visualization. Emotional LED lighting control system is suggested to help audiences enhance the musical experience. In the system, emotion in music is recognized by a proposed algorithm using a dimensional approach. The algorithm used a method of music emotion variation detection to overcome some weaknesses of Thayer's model in detecting emotion in a one-second music segment. In addition, IRI color model is combined with Thayer's model to determine LED light colors corresponding to 36 different music emotions. They are represented on LED lighting control system through colors and animations. The accuracy of music emotion visualization achieved to over 60%.

• Science of Emotion and Sensibility
• /
• v.15 no.2
• /
• pp.283-296
• /
• 2012

The goal of this study is to determine the social emotion model as an emotion sharing relationship and information sharing relationship based on the user's relations at social networking services. 26 social emotions were extracted by verification of compliance among 92 different emotions collected from the literature survey. The survey on the 26 emotion words was verified to the similarity of social relation types to the Likert 7-points scale. The principal component analysis of the survey data determined 12 representative social emotions in the emotion sharing relation and 13 representative social emotions in the information sharing relation. Multidimensional scaling developed the two-dimensional social emotion model of emotion sharing relation and of information sharing relation based on online communication environment. Meanwhile, insignificant factors in the suggest social emotion models were removed by the structural equation modeling analysis, statistically. The test result of validity analysis demonstrated the fitness of social emotion models at emotion sharing relationships (CFI: .887, TLI: .885, RMSEA: .094), social emotion model of information sharing relationships (CFI: .917, TLI: .900, RMSEA : 0.050). In conclusion, this study presents two different social emotion models based on two different relation types. The findings of this study will provide not only a reference of evaluating social emotions in designing social networking services but also a direction of improving social emotions.

• Journal of Korea Multimedia Society
• /
• v.12 no.8
• /
• pp.1155-1163
• /
• 2009

This paper proposes an automatic facial expression generation system of vector graphic character using gaussian process model. Proposed method extracts the main feature vectors from twenty-six facial data of character redefined based on Russell's internal emotion state. Also by using new gaussian process model, SGPLVM, we find low-dimensional feature data from extracted high-dimensional feature vectors, and learn probability distribution function (PDF). All parameters of PDF are estimated by maximization the likelihood of learned expression data, and these are used to select wanted facial expressions on two-dimensional space in real time. As a result of simulation, we confirm that proposed facial expression generation tool is working in the small facial expression datasets and can generate various facial expressions without prior knowledge about relation between facial expression and emotion.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.165-176
• /
• 2018

With the development of information and communication technology, the possibility of utilizing 3D virtual reality in education has been sought. Especially, the screen composition in the virtual reality has the possibility of inducing the emotion of the user which may affect the learning. However, there is little research on what aspects of the screen can cause emotions. The purpose of this study is to analyze the user's perception of screen components inducing emotion in virtual reality learning environment. Using Multi Dimensional Scaling (MDS), the user's perception of the main screen in a representative virtual reality learning environment platform was investigated. As a result, the dimension of depth on the screen and the dynamics of the avatar related to the movement were confirmed. This study is meaningful to explore technical variables that can induce emotions among screen elements in virtual reality contents.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.6
• /
• pp.284-290
• /
• 2024

Speech emotion recognition (SER) is a technique that is used to analyze the speaker's voice patterns, including vibration, intensity, and tone, to determine their emotional state. There has been an increase in interest in artificial intelligence (AI) techniques, which are now widely used in medicine, education, industry, and the military. Nevertheless, existing researchers have attained impressive results by utilizing acted-out speech from skilled actors in a controlled environment for various scenarios. In particular, there is a mismatch between acted and spontaneous speech since acted speech includes more explicit emotional expressions than spontaneous speech. For this reason, spontaneous speech-emotion recognition remains a challenging task. This paper aims to conduct emotion recognition and improve performance using spontaneous speech data. To this end, we implement deep learning-based speech emotion recognition using the VGG (Visual Geometry Group) after converting 1-dimensional audio signals into a 2-dimensional spectrogram image. The experimental evaluations are performed on the Korean spontaneous emotional speech database from AI-Hub, consisting of 7 emotions, i.e., joy, love, anger, fear, sadness, surprise, and neutral. As a result, we achieved an average accuracy of 83.5% and 73.0% for adults and young people using a time-frequency 2-dimension spectrogram, respectively. In conclusion, our findings demonstrated that the suggested framework outperformed current state-of-the-art techniques for spontaneous speech and showed a promising performance despite the difficulty in quantifying spontaneous speech emotional expression.

• Science of Emotion and Sensibility
• /
• v.2 no.1
• /
• pp.43-52
• /
• 1999

정상 성인의 경우, 감성 개념의 내적 차원 구조는 쾌/불쾌 차원과 각성/수면 차원이라는 이차원 공간상에 원형으로 분포되는 양상을 보인다고 아려져 왔다. 본 연구에서는 이와 같은 이차원 구조가 얼마나 보편적이고 일관적인가를 알아보고자 했다. 이를 위해 연구 1에서는 교차타당화를 통해 이차원 구조의 안정성을 검토했으며, 연구 2에서는 다양한 성격 특질을 지닌 집단을 선발하여 22개 단어를 짝지은 231개 쌍에 대해 유사성 평정을 시비고 다차원분석법(MDS)으로 분석한 후 이를 연구 1의 대학생 집단의 결과와 비교하였다. 연구 3에서는 낮은 발달 수준에 있는 아동 및 청소년 집단을 대상으로 연구 2와 동일한 절차를 통해 분석하였다. 연구 1, 2, 3 모두 감성 개념의 이차원 구조가 동일한 것으로 나타나 매우 안정적인 구조라는 것이 입증되었으며, 다만 아동 및 청소년의 경우 1 차원의 설명량이 정상 성인 집단보다 크고 각성/수면 차원의 설명량은 더 적었다. 이러한 결과를 통해 이차원 구조의 안정성과 일반화 가능성, 제한점, 그리고 집단의 독특한 특성을 반영하는데 이 구조적 틀을 적용할 수 있는 유용성에 대해 논의하였다.

• Science of Emotion and Sensibility
• /
• v.7 no.3
• /
• pp.51-56
• /
• 2004

Arousal level has been well defined by autonomic responses. However, entire emotion including both valence and arousal level is often questioned to be completely described by only autonomic responses. This study is to find the autonomic physiological parameters which were used emotion evaluation, 15 undergraduate students were asked to watch eight video clips from diverse movies and comedy shows for experiencing emotions. The subjectively experienced emotion were grouped by three factors. Two dimensional emotion model having the pleasant-unpleasant and arousal-non arousal factors were mapped with three physiological responses(GSR, PPG, SKT). The results may suggest that PPG and GSR may be used as arousal index while SKT may pleasant index. And the complex relation of physiological responses to emotional experiences are discussed.