• Journal of the Korea Computer Graphics Society
• /
• v.29 no.4
• /
• pp.17-24
• /
• 2023

In computer graphics and computer vision research and its applications, various representations of 3D objects, such as point clouds, voxels, or triangular meshes, are used depending on the purpose. The need for animating characters using these representations is also growing. In a typical animation pipeline called skeletal animation, "skinning weight painting" is required to determine how joints influence a vertex on the character's skin. In this paper, we introduce a neural network for automatically performing skinning weight painting for characters represented in various formats. We utilize signed distance fields (SDF) to handle different representations and employ graph neural networks and multi-layer perceptrons to predict the skinning weights for a given point.

• Journal of KIISE:Software and Applications
• /
• v.35 no.9
• /
• pp.572-587
• /
• 2008

The Self-organizing Feature Map(SOFM) that is one of unsupervised neural networks is a very powerful tool for data clustering and visualization in high-dimensional data sets. Although the SOFM has been applied in many engineering problems, it needs to cluster similar weights into one class on the trained SOFM as a post-processing, which is manually performed in many cases. The traditional clustering algorithms, such as t-means, on the trained SOFM however do not yield satisfactory results, especially when clusters have arbitrary shapes. This paper proposes automatic clustering on trained SOFM, which can deal with arbitrary cluster shapes and be globally optimized by graph cuts. When using the graph cuts, the graph must have two additional vertices, called terminals, and weights between the terminals and vertices of the graph are generally set based on data manually obtained by users. The Proposed method automatically sets the weights based on mode-seeking on a distance matrix. Experimental results demonstrated the effectiveness of the proposed method in texture segmentation. In the experimental results, the proposed method improved precision rates compared with previous traditional clustering algorithm, as the method can deal with arbitrary cluster shapes based on the graph-theoretic clustering.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.45 no.6
• /
• pp.978-985
• /
• 2021

This study aims to propose a prediction model for the drape coefficient using artificial neural networks and to analyze the nonlinear relationship between the drape properties and physical properties of fabrics. The study validates the significance of each factor affecting the fabric drape through multiple linear regression analysis with a sample size of 573. The analysis constructs a model with an adjusted R² of 77.6%. Seven main factors affect the drape coefficient: Grammage, extruded length values for warp and weft (m_warp, m_weft), coefficients of quadratic terms in the tensile-force quadratic graph in the warp, weft, and bias directions (c_warp, c_weft, c_bias), and force required for 1% tension in the warp direction (f_warp). Finally, an artificial neural network was created using seven selected factors. The performance was examined by increasing the number of hidden neurons, and the most suitable number of hidden neurons was found to be 8. The mean squared error was .052, and the correlation coefficient was .863, confirming a satisfactory model. The developed artificial neural network model can be used for engineering and high-quality clothing design. It is expected to provide essential data for clothing appearance, such as the fabric drape.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1469-1476
• /
• 2022

Recently, many researches have been conducted to estimate blood pressure using ECG(Electrocardiogram) and PPG(Photoplentysmography) signals. In this paper, we designed and implemented a mobile system to monitor blood pressure in real time by using 1-D convolutional neural networks. The proposed model consists of deep 11 layers which can learn to extract various features of ECG and PPG signals. The simulation results show that the more the number of convolutional kernels the learned neural network has, the more detailed characteristics of ECG and PPG signals resulted in better performance with reduced mean square error compared to linear regression model. With receiving measurement signals from wearable ECG and PPG sensor devices attached to the body, the developed system receives measurement data transmitted through Bluetooth communication from the devices, estimates systolic and diastolic blood pressure values using a learned model and displays its graph in real time.

• Journal of Multimedia Information System
• /
• v.9 no.3
• /
• pp.177-182
• /
• 2022

A deep neural network (DNN) includes variables whose values keep on changing with the training process until it reaches the final point of convergence. These variables are the co-efficient of a polynomial expression to relate to the feature extraction process. In general, DNNs work in multiple 'dimensions' depending upon the number of channels and batches accounted for training. However, after the execution of feature extraction and before entering the SoftMax or other classifier, there is a conversion of features from multiple N-dimensions to a single vector form, where 'N' represents the number of activation channels. This usually happens in a Fully connected layer (FCL) or a dense layer. This reduced 2D feature is the subject of study for our analysis. For this, we have used the FCL, so the trained weights of this FCL will be used for the weight-class correlation analysis. The popular DNN models selected for our study are ResNet-101, VGG-19, and GoogleNet. These models' weights are directly used for fine-tuning (with all trained weights initially transferred) and scratch trained (with no weights transferred). Then the comparison is done by plotting the graph of feature distribution and the final FCL weights.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.5
• /
• pp.911-919
• /
• 2011

In recent medical analysis, it becomes more important to looking for risk factors related to mental illness. If we find and identify their relevant characteristics of the risk factors, the disease can be prevented in advance. Moreover, the study can be helpful to medical development. These kinds of studies of risk factors for mental illness have mainly been discussed by using the logistic regression model. However in this paper, data mining techniques such as CART, C5.0, logistic, neural networks and Bayesian network were used to search for the risk factors. The Bayesian network of the above data mining methods was selected as most optimal model by applying delirium data. Then, Bayesian network analysis was used to find risk factors and the relationship between the risk factors are identified through a radial graph.

• Journal of KIISE:Software and Applications
• /
• v.31 no.12
• /
• pp.1581-1590
• /
• 2004

We present a robust off-line method recognizing handwritten digits by only using stroke direction codes as a feature of handwritten digits. This method makes general 8-direction codes for an input digit and then has the multi-layered neural networks learn them and recognize each digit. The 8-direction codes are made of the thinned results of each digit through SOG*(Improved Self-Organizing Graph). And the usage of these codes simplifies the complex steps processing at least two features of the existing methods. The experimental result shows that the recognition rates of this method are constantly better than 98.85% for any images in all digit databases.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.5
• /
• pp.73-79
• /
• 2021

Recently, in the field of computer animation, a method for generating motion using deep learning has been studied away from conventional finite-state machines or graph-based methods. The expressiveness of the network required for learning motions is more influenced by the diversity of motion contained in it than by the simple length of motion to be learned. This study aims to find an efficient network structure when the types of motions to be learned are diverse. In this paper, we train and compare three types of networks: basic fully-connected structure, mixture of experts structure that uses multiple fully-connected layers in parallel, recurrent neural network which is widely used to deal with seq2seq, and transformer structure used for sequence-type data processing in the natural language processing field.

• Journal of Internet Computing and Services
• /
• v.22 no.5
• /
• pp.27-33
• /
• 2021

With the broad adoption of the Internet of Things (IoT) in a variety of scenarios and application services, management and orchestration entities require upgrading the traditional architecture and develop intelligent models with ultra-reliable methods. In a heterogeneous network environment, mission-critical IoT applications are significant to consider. With erroneous priorities and high failure rates, catastrophic losses in terms of human lives, great business assets, and privacy leakage will occur in emergent scenarios. In this paper, an efficient resource slicing scheme for optimizing federated learning in software-defined IoT (SDIoT) is proposed. The decentralized support vector regression (SVR) based controllers predict the IoT slices via packet inspection data during peak hour central congestion to achieve a time-sensitive condition. In off-peak hour intervals, a centralized deep neural networks (DNN) model is used within computation-intensive aspects on fine-grained slicing and remodified decentralized controller outputs. With known slice and prioritization, federated learning communications iteratively process through the adjusted resources by virtual network functions forwarding graph (VNFFG) descriptor set up in software-defined networking (SDN) and network functions virtualization (NFV) enabled architecture. To demonstrate the theoretical approach, Mininet emulator was conducted to evaluate between reference and proposed schemes by capturing the key Quality of Service (QoS) performance metrics.

• Proceedings of the KIEE Conference
• /
• 1997.11a
• /
• pp.203-205
• /
• 1997

In this paper, we present a useful method of estimating earth resistivity using BP algorithm in Neural-Networks. From this method, equivalent earth resistivity(EER) can be obtained directly using training data composed of field-measured apparent resistivity and probe distance. This approach can reduce errors which is conventionally raised from manual operation of calculating EER. To evaluate its accuracy and convenience, the result of proposed method is compared to that of conventional methods, graphical($\rho$-a graph) and numerical(CDEGS program), respectively.