• Journal of the Korea Computer Graphics Society
• /
• v.8 no.4
• /
• pp.17-25
• /
• 2002

In this paper we propose a new efficient muscle modeling for creating 3d facial models. It improves the existing muscle-based facial modeling method. We present the facial muscle action prediction map as a new method of muscle modeling. We also suggest a new face deformation method that improves the existing method which controls each muscle's condition separately.

• 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.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2004.10a
• /
• pp.382-387
• /
• 2004

As the increasing needs in the industrial filed, many studies for the 3D CAD system are carried out. There are two types of 3D CAD system. One is manifold modeler, the other is non-manifold modeler. In the manifold modeler only 3D objects can be modeled. In the non-manifold modeler 3D, 2D, 1D, and 0D objects can be modeled in a unified data structure. Recently there are many studies on the non-manifold modeler. Most of them are focused on finding unknown topological entities and representing all kinds of topological entities found. In this paper, efficient data structure is selected. The boundary information on a face and an edge is included in this data structure. The boundary information on a vertex is excluded considering the frequency of usage. Because the disk cycle information is not required in most case of modeling. It is compact. It stores essential non-manifold information such as loop cycle and radial cycle. A suitable Euler-Poincare equation is studied and selected. Using the efficient data structure and the selected Euler-Poincare equation, 18 basic Euler operators are implemented. Several 3D models are created using the implemented modeler. A non-manifold modeling can be carried out using the implemented 3D CAD system. The results of this paper could be used in the further studies such as an implementation of Boolean operators, and a translation of 2D CAD drawings to 3D models.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.153.6-153
• /
• 2001

3-Dimentional 3D digitization of the human is a technology that is still relatively new. There are present uses such as radiotherapy, identification systems and commercial uses and potential future applications. In this paper, we analyzed and experimented to determine the easiest and most efficient method, which would give us the most accurate results. We also constructed a database of realistic expressions and high quality human heads. We scanned people´s heads and facial expressions in 3D using a Minolta Vivid 700 scanner, then edited the models obtained on a Silicon Graphics workstation. Research was done into the present and potential uses of the 3D digitized models of the human head and we develop ideas for ...

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.1481-1498
• /
• 2007

As railroad industry face the new Renaissance era, effective and efficient maintenance methods for rolling stock operation are required with advanced railroad technology. All kinds of railroad systems such as high speed long distance train, metropolitan mass transit and light rail require systematic maintenance technology in order to maintain the safe railroad operation. Simulation models for detailed operations of the sample maintenance center are developed. In this study, standard maintenance procedures, layout, equipments and number of workers of Siheung Metropolitan Railroad Maintenance Rolling Stock Depot are considered. The proposed simulation models are developed using simulation package ARENA. Three simulation analysis using the developed simulation model are done. First, the bottleneck operation is identified. Second, the relationship between maintenance center size, number of workers and cycle time is analyzed. Lastly, the scheduling performances between PERT/CPM and Critical Chain Project Management(CCPM) are compared.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.539-554
• /
• 2018

Effective interaction between user and device is considered an important ability of IoT devices. For some applications, it is necessary to recognize human facial expressions in real time and make accurate judgments in order to respond to situations correctly. Therefore, many researches on facial image analysis have been preceded in order to construct a more accurate and faster recognition system. In this study, we constructed an automatic recognition system for facial expressions through two steps - a facial recognition step and a classification step. We compared various models with different sets of data with pixel information, landmark coordinates, Euclidean distances among landmark points, and arctangent angles. We found a fast and efficient prediction model with only 30 principal components of face landmark information. We applied several prediction models, that included linear discriminant analysis (LDA), random forests, support vector machine (SVM), and bagging; consequently, an SVM model gives the best result. The LDA model gives the second best prediction accuracy but it can fit and predict data faster than SVM and other methods. Finally, we compared our method to Microsoft Azure Emotion API and Convolution Neural Network (CNN). Our method gives a very competitive result.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.3 s.303
• /
• pp.63-72
• /
• 2005

In this paper, we propose a deformation technique based on Radial Basis Function (RBF) and a blending technique combining the deformed facial component with the original face for a Virtual Aesthetic Surgery (VAS) system. The deformation technique needs the smoothness and the accuracy to deform the fluid facial components and also needs the locality not to affect or distort the rest of the facial components besides the deformation region. To satisfy these deformation characteristics, The VAS System computes the degree of deformation of lattice cells using RBF based on a Free-Form Deformation (FFD) model. The deformation error is compensated by the coefficients of mapping function, which is recursively solved by the Singular Value Decomposition (SVD) technique using SSE (Sum of Squared Error) between the deformed control points and target control points on base curves. The deformed facial component is blended with an original face using a blending ratio that is computed by the Euclidean distance transform. An experimental result shows that the proposed deformation and blending techniques are very efficient in terms of accuracy and distortion.

• Journal of Broadcast Engineering
• /
• v.26 no.5
• /
• pp.608-620
• /
• 2021

Recent deep learning-based face super-resolution (FSR) works showed significant performances by utilizing facial prior knowledge such as facial landmark and dictionary that reflects structural or semantic characteristics of the human face. However, most of these methods require additional processing time and memory. To solve this issue, this paper propose an efficient FSR models using knowledge distillation techniques. The intermediate features of teacher network which contains dictionary information based on major face regions are transferred to the student through adversarial multi-scale features distillation. Experimental results show that the proposed model is superior to other SR methods, and its effectiveness compare to teacher model.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.93-101
• /
• 2024

Accurate field crop classification is essential for various agricultural applications, yet existing methods face challenges due to diverse crop types and complex field conditions. This study aimed to address these issues by combining support vector machine (SVM) models with multi-seasonal unmanned aerial vehicle (UAV) images, texture information extracted from Gray Level Co-occurrence Matrix (GLCM), and RGB spectral data. Twelve high-resolution UAV image captures spanned March-October 2021, while field surveys on three dates provided ground truth data. We focused on data from August (-A), September (-S), and October (-O) images and trained four support vector classifier (SVC) models (SVC-A, SVC-S, SVC-O, SVC-AS) using visual bands and eight GLCM features. Farm maps provided by the Ministry of Agriculture, Food and Rural Affairs proved efficient for open-field crop identification and served as a reference for accuracy comparison. Our analysis showcased the significant impact of hyperparameter tuning (C and gamma) on SVM model performance, requiring careful optimization for each scenario. Importantly, we identified models exhibiting distinct high-accuracy zones, with SVC-O trained on October data achieving the highest overall and individual crop classification accuracy. This success likely stems from its ability to capture distinct texture information from mature crops.Incorporating GLCM features proved highly effective for all models,significantly boosting classification accuracy.Among these features, homogeneity, entropy, and correlation consistently demonstrated the most impactful contribution. However, balancing accuracy with computational efficiency and feature selection remains crucial for practical application. Performance analysis revealed that SVC-O achieved exceptional results in overall and individual crop classification, while soybeans and rice were consistently classified well by all models. Challenges were encountered with cabbage due to its early growth stage and low field cover density. The study demonstrates the potential of utilizing farm maps and GLCM features in conjunction with SVM models for accurate field crop classification. Careful parameter tuning and model selection based on specific scenarios are key for optimizing performance in real-world applications.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.317-326
• /
• 2024

Biometric recognition is a technology that determines whether a person is identified by extracting information on a person's biometric and behavioral characteristics with a specific device. Cyber threats such as forgery, duplication, and hacking of biometric characteristics are increasing in the field of biometrics. In response, the security system is strengthened and complex, and it is becoming difficult for individuals to use. To this end, multiple biometric models are being studied. Existing studies have suggested feature fusion methods, but comparisons between feature fusion methods are insufficient. Therefore, in this paper, we compared and evaluated the fusion method of multiple biometric models using fingerprint, face, and iris images. VGG-16, ResNet-50, EfficientNet-B1, EfficientNet-B4, EfficientNet-B7, and Inception-v3 were used for feature extraction, and the fusion methods of 'Sensor-Level', 'Feature-Level', 'Score-Level', and 'Rank-Level' were compared and evaluated for feature fusion. As a result of the comparative evaluation, the EfficientNet-B7 model showed 98.51% accuracy and high stability in the 'Feature-Level' fusion method. However, because the EfficietnNet-B7 model is large in size, model lightweight studies are needed for biocharacteristic fusion.