• The Journal of Korea Robotics Society
• /
• v.8 no.4
• /
• pp.266-272
• /
• 2013

Facial feature extraction and tracking are essential steps in human-robot-interaction (HRI) field such as face recognition, gaze estimation, and emotion recognition. Active shape model (ASM) is one of the successful generative models that extract the facial features. However, applying only ASM is not adequate for modeling a face in actual applications, because positions of facial features are unstably extracted due to limitation of the number of iterations in the ASM fitting algorithm. The unaccurate positions of facial features decrease the performance of the emotion recognition. In this paper, we propose real-time facial feature extraction and tracking framework using ASM and LK optical flow for emotion recognition. LK optical flow is desirable to estimate time-varying geometric parameters in sequential face images. In addition, we introduce a straightforward method to avoid tracking failure caused by partial occlusions that can be a serious problem for tracking based algorithm. Emotion recognition experiments with k-NN and SVM classifier shows over 95% classification accuracy for three emotions: "joy", "anger", and "disgust".

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.3
• /
• pp.1390-1403
• /
• 2016

Facial expression recognition (FER) plays a very significant role in computer vision, pattern recognition, and image processing applications such as human computer interaction as it provides sufficient information about emotions of people. For video-based facial expression recognition, depth cameras can be better candidates over RGB cameras as a person's face cannot be easily recognized from distance-based depth videos hence depth cameras also resolve some privacy issues that can arise using RGB faces. A good FER system is very much reliant on the extraction of robust features as well as recognition engine. In this work, an efficient novel approach is proposed to recognize some facial expressions from time-sequential depth videos. First of all, efficient Local Binary Pattern (LBP) features are obtained from the time-sequential depth faces that are further classified by Generalized Discriminant Analysis (GDA) to make the features more robust and finally, the LBP-GDA features are fed into Hidden Markov Models (HMMs) to train and recognize different facial expressions successfully. The depth information-based proposed facial expression recognition approach is compared to the conventional approaches such as Principal Component Analysis (PCA), Independent Component Analysis (ICA), and Linear Discriminant Analysis (LDA) where the proposed one outperforms others by obtaining better recognition rates.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.35 no.4
• /
• pp.199-204
• /
• 2009

Purpose: The purpose of this study is to investigate the epithelial-mesenchymal transition (EMT) induced by Snail transcription factor and Snail-transfected in vivo tumors with histopathological features. Materials and methods: We induced in vivo xenografted tumorigenesis in the oral vestibules of nude mice by a Snail transfected HaCaT cell line and investigated morphological and immunohistochemical features in Snail expressive tumors. Results: We identified tumor masses in 14 out of 15 nude mice in the HaCaT-Snail cell inoculation group, but no tumors were present in any of the HaCaT cell inoculation group. Induced tumors showed features of poorly differentiated carcinoma with invasion to neighboring muscles and bones. The HaCaT-Snail tumors showed decreased expressions of E-cadherin and cytokeratin, but showed increased expressions of vimentin and N-cadherin. Discussion: The Snail transfected xenograft can improve productivity of malignant tumors, show various histopathological features including invasive growth, and aid in the investigation of tumor progression and the interaction with surrounding tissues.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.1
• /
• pp.43.1-43.1
• /
• 2014

It has been suggested that only the most luminous AGNs ($L{\geq}$ [10] $^{45}L_{\odot}$ ) are triggered by galaxy mergers, while less luminous AGNs (L~ [10] $^{43}L_{\odot}$) are driven by other internal processes. Lack of merging features in low luminosity AGN host galaxies has been a main argument against the idea of merger triggering of low luminosity AGNs, but merging, especially a rather minor one, might still have played an important role in low luminosity AGNs since minor merging features in low luminosity are more difficult to identify than major merging features. Using SNUCAM on the 1.5m telescope at Madanak observatory, we obtained deep images of NGC 7743 which is a barred spiral galaxy classified as a Seyfert 2 AGN with a low bolometric luminosity of $5{\times}$ [10] $^{42}L_{\odot}$. Surprisingly, we newly discovered merging features around the galaxy, which indicate past merging activity on the galaxy. This example indicates the merging fraction of low luminosity AGNs may be much higher than previously thought, hinting the importance of galaxy merger even in low luminosity AGN.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2768-2787
• /
• 2023

As an important research direction of the application of computer science in the medical field, the automatic generation technology of radiology report has attracted wide attention in the academic community. Because the proportion of normal regions in radiology images is much larger than that of abnormal regions, words describing diseases are often masked by other words, resulting in significant feature loss during the calculation process, which affects the quality of generated reports. In addition, the huge difference between visual features and semantic features causes traditional multi-modal fusion method to fail to generate long narrative structures consisting of multiple sentences, which are required for medical reports. To address these challenges, we propose a multi-feature optimization Transformer (MFOT) for generating radiology reports. In detail, a multi-dimensional mapping attention (MDMA) module is designed to encode the visual grid features from different dimensions to reduce the loss of primary features in the encoding process; a feature pre-fusion (FP) module is constructed to enhance the interaction ability between multi-modal features, so as to generate a reasonably structured radiology report; a detail enhanced attention (DEA) module is proposed to enhance the extraction and utilization of key features and reduce the loss of key features. In conclusion, we evaluate the performance of our proposed model against prevailing mainstream models by utilizing widely-recognized radiology report datasets, namely IU X-Ray and MIMIC-CXR. The experimental outcomes demonstrate that our model achieves SOTA performance on both datasets, compared with the base model, the average improvement of six key indicators is 19.9% and 18.0% respectively. These findings substantiate the efficacy of our model in the domain of automated radiology report generation.

• Journal of Korea Game Society
• /
• v.11 no.4
• /
• pp.37-48
• /
• 2011

As people increasingly play online games, numerous new features have been proposed to increase players' log-on time at online gaming sites. However, few studies have investigated why people continue to play certain online games. This research would verify that enjoyment experience could be explained by the conceptual framework. In the first, this study results indicate that customers would show a higher level of loyalty if they had an optimal experience with the games. The state of flow was felt when players were aware of opportunities for personal interaction and social interaction. The personal interaction could be motivated either to achieve the high cognitive performance or by providing the equity of distribution of the objective performance in order to examine players' cognitive performance; the social interaction can be motivated either to enhance a high level of self-esteem of player or to achieve the positive reputation in order to evaluate their self-esteem. This finding can answer the questions of what enjoyment experience is and why players are repeatedly playing specific online games.

• Journal of the Korean Chemical Society
• /
• v.61 no.5
• /
• pp.251-262
• /
• 2017

The purpose of this study was to investigate characteristics of interaction in mentoring conversations and to examine how the interaction features change as mentor teachers have more mentoring experiences. Participants of this study were three mentor who have over 17 years' teaching experience and six beginning science teachers. For this study, one-to-one mentoring dialogue recordings and transcripts were collected and the dialogues were analyzed by utilizing an analytical framework of interaction. the result of analyzing characteristics of mentors' interaction shows that mentors used simple questions and support the most when they started mentoring conversation. the change of characteristics of mentor's interactions indicates three mentors tended to use more thought-provoking questions in the $2^{nd}$ year mentoring than in the $1^{st}$ year and as a result of it mentee's reflection and reflective practices were increased. Through mentors' interview, the mentors could have the opportunity to reflect their own mentoring and this means mentors' self reflection was provoked by means of the mentoring program.

• 한국경영정보학회:학술대회논문집
• /
• 2007.06a
• /
• pp.640-645
• /
• 2007

In theese day, games are often referred as the most potential media to have artistic and entertaining features of digital technology, In this study, used to derive the factors of enjoyment and immersion that users experience from interaction and narrative levels in games were derived The framework of interaction and narrative levels in games was adopted from Marie-Laure Ryan's(2002) and Hyewon Han's(2005) study. The framework was used as the basic instrument for focus group interviews. The result of this study provides a practical model for game-design.

• Wind and Structures
• /
• v.5 no.2_3_4
• /
• pp.101-114
• /
• 2002

Both a Finite Volume and a Discrete Vortex technique to solve the unsteady Navier-Stokes equations have been employed to study the air flow around long-span bridge decks. The implementation and calibration of both methods is described alongside a quasi-3D extension added to the DVM solver. Applications to the wind engineering of bridge decks include flow simulations at different angles of attack, calculation of aerodynamic derivatives and fluid-structure interaction analyses. These are being presented and their specific features described. If a numerical method shall be employed in a practical design environment, it is judged not only by its accuracy but also by factors like versatility, computational cost and ease of use. Conclusions are drawn from the analyses to address the question of whether computer simulations can be practical design tools for the wind engineering of bridge decks.

• 한국연소학회:학술대회논문집
• /
• 2012.11a
• /
• pp.85-87
• /
• 2012

We present a numerical investigation on gaseous (ethylene-air mixture) detonation in the elastoplastical metal tubes to understand the wall effects associated with the developing detonation instability. The acoustic disturbances originating from the rapidly expanding tube walls reach the detonating flame surface, thereby causing flame distortions and total energy losses. The compressible Navier-Stokes equations with equation of state for gas and elasto-plastic deformation field equations for inert tubes are solved simultaneously to understand the complex multi-material interaction in the rapidly expanding gas pipe. In order to track governing variables across the material interface, we use the hybrid particle level-set and ghost fluid methods to precisely estimate the interfacial quantities. Features observed from the deforming (thin) tube show substantially different behavior when a detonation propagates in the rigid (thick) tube with no acoustically responding wall conditions.