• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.186-203
• /
• 2020

Image-based Virtual Try-On Systems are among the most potential solution for virtual fitting which tries on a target clothes into a model person image and thus have attracted considerable research efforts. In many cases, current solutions for those fails in achieving naturally looking virtual fitted image where a target clothes is transferred into the body area of a model person of any shape and pose while keeping clothes context like texture, text, logo without distortion and artifacts. In this paper, we propose a new improved image-based virtual try-on network system based on keypoints, which we name as KP-VTON. The proposed KP-VTON first detects keypoints in the target clothes and reliably predicts keypoints in the clothes of a model person image by utilizing a dense human pose estimation. Then, through TPS transformation calculated by utilizing the keypoints as control points, the warped target clothes image, which is matched into the body area for wearing the target clothes, is obtained. Finally, a new try-on module adopting Attention U-Net is applied to handle more detailed synthesis of virtual fitted image. Extensive experiments on a well-known dataset show that the proposed KP-VTON performs better the state-of-the-art virtual try-on systems.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.1
• /
• pp.79-83
• /
• 2021

The expiration date of high energetic materials(HEM), such as HMX, RDX, TNT, is important. If the expiration date is violated, the expected specification of HEM would not be satisfied which may cause a different conclusion in an urgent situation. As a result, this HEM should maintain fresh conditions which cause the accumulation of waste HEM. If HEM is landfilled during demilitarization, the impact on living organizations is serious. Additionally, landfilling HEM has a possibility of explosion. In this research, the process flow diagram of the demilitarization gas treatment process was simulated while satisfying the law of the environment in Korea. After validation of simulation, it was optimized thermodynamically using Heat Exchanger Network Synthesis(HENs). This study is expected to enhance the energy efficiency of the original facility by suggesting developed designs. This research was supported by Agency of Defense Development NE32 Korea. Thanks to Agency of Defense Development, Korea

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1997.06a
• /
• pp.27-34
• /
• 1997

In this study , we tried to test the classification performance of a neural netow and thereby to examine its applicability to the signals distorted by a shallow water einvironment . We conducted an acoustic experiment iin a shallow sea near Pohang, Korea in which water depth is about 60m. The signals, on which the network has been tested, is ilinear frequency modulated ones centered on one of the frequencies, 200, 400, 600 and 800 Hz, each being swept up or down with bandwidth 100Hz. we considered two transforms, STFT(short-time Fourier transform) and PWVD (pseudo Wigner-Ville distribution), form which power spectra were derived. The training signals were simulated using an acoutic model based on the Fourier synthesis scheme. When the network has been trained on the measured signals of center frequency 600Hz,it gave a little better results than that trained onthe simulated . With the center frequencies varied, the overall performance reached over 90% except one case of center frequency 800Hz. With the feature extraction techniques(STFT and PWVD) varied,the network showed performance comparable to each other . In conclusion , the signals which have been simulated with water depth were successully applied to training a neural network, and the trained network performed well in classifying the signals distorted by a surrounding environment and corrupted by noise.

• Korean Chemical Engineering Research
• /
• v.44 no.1
• /
• pp.97-105
• /
• 2006

In order to investigate the effect of inhibitors on L-threonine biosynthesis in Escherichia coli, we have constructed a metabolic network model of amino acid biosynthesis from L-aspartate to L-threonine by using available informations from literatures and databases. In the model, the effects of inhibitors on the biosynthesis of L-threonine was included as an appropriate mathematical form. For simulation study, we used initial values as L-aspartate 5 mM, ATP 5 mM, NADPH 2 mM, and observed the concentration changes of intermediate metabolites over concentration changes of respective inhibitors. As a result, we found that concentrations of intermediate metabolites were not significantly changed over concentration changes of L-lysine, L-methionine, and L-glutamate. But, there were considerable changes of intermediates over concentration changes of L-serine, L-cysteine, and L-threonine, which can be considered as essential effectors on L-threonine synthesis. Contrary, the synthesis of L-threonine seems to be not related to the amounts of L-aspartate, and inversely proportional to the accumulated amount of D,L-aspartic ${\beta}$-semialdehyde.

• Structural Engineering and Mechanics
• /
• v.32 no.1
• /
• pp.71-94
• /
• 2009

In this paper a procedure for Monte Carlo simulation of univariate stationary stochastic processes with the aid of neural networks is presented. Neural networks operate model-free and, thus, circumvent the need of specifying a priori statistical properties of the process, as needed traditionally. This is particularly advantageous when only limited data are available. A neural network can capture the "pattern" of a short observed time series. Afterwards, it can directly generate stochastic process realizations which capture the properties of the underlying data. In the present study a simple feed-forward network with focused time-memory is utilized. The proposed procedure is demonstrated by examples of Monte Carlo simulation, by synthesis of future values of an initially short single process record.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.301-303
• /
• 2000

This paper presents Kohonen neural network and wavelet transform analysis based technique for industrial peak load forecasting for the purpose of peak demand control. Firstly, one year of historical load data were sorted and clustered into several groups using Kohonen neural network and then wavelet transforms are adopted using the Biorthogonal mother wavelet in order to forecast the peak load of one hour ahead. The 5-level decomposition of the daily industrial load curve is implemented to consider the weather sensitive component of loads effectively. The wavelet coefficients associated with certain frequency and time localization is adjusted using the conventional multiple regression method and the components are reconstructed to predict the final loads through a six-scale synthesis technique.

• Proceedings of the KIEE Conference
• /
• 1999.11b
• /
• pp.239-241
• /
• 1999

This paper proposes a novel wavelet transform and Kohonen neural network based technique for short-time load forecasting of power systems. Firstly. Kohonen Self-organizing map(KSOM) is applied to classify the loads and then the Daubechies D2, D4 and D10 wavelet transforms are adopted in order to forecast the short-term loads. The wavelet coefficients associated with certain frequency and time localisation are adjusted using the conventional multiple regression method and then reconstructed in order to forecast the final loads through a four-scale synthesis technique. The outcome of the study clearly indicates that the proposed composite model of Kohonen neural network and wavelet transform approach can be used as an attractive and effective means for short-term load forecasting.

• Journal of Internet Computing and Services
• /
• v.19 no.4
• /
• pp.35-44
• /
• 2018

Monitoring through Synthesis Aperture Radar (SAR) is responsible for marine safety from floating icebergs. However, there are limits to distinguishing between icebergs and ships in SAR images. Convolutional Neural Network (CNN) is used to distinguish the iceberg from the ship. The goal of this paper is to increase the accuracy of identifying icebergs from SAR images. The metrics for performance evaluation uses the log loss. The two-layer CNN model proposed in research of C.Bentes et al.[1] is used as a benchmark model and compared with the four-layer CNN model using data augmentation. Finally, the performance of the final CNN model using the VGG-16 pre-trained model is compared with the previous model. This paper shows how to improve the benchmark model and propose the final CNN model.

• Journal of the Korea Safety Management & Science
• /
• v.8 no.4
• /
• pp.167-180
• /
• 2006

The ANP(Analytic Network Process), though based on the AHP(Analytic Hierarchy Process), is a system for the analysis, synthesis, and justification of complex decisions with the capability to model non-linear relations between the elements. ANP allows the decision makers to leap beyond the traditional hierarchy to the interdependent environment of network modeling. The ANP is designed for problems characterized by the added complexity of interdependencies such as feedback and dependencies among problem elements. Using a network approach makes it possible to represent and analyze interactions, incorporate non-linear relations between the elements, and synthesize mutual effects by a single logical procedure. This study intends to evaluate the contribution of technology in intangible assets by the AHP and ANP.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.3
• /
• pp.149-154
• /
• 2023

This paper presents a novel approach to multi-focus image fusion using light field cameras. The proposed neural network, LFFCNN (Light Field Focus Convolutional Neural Network), is composed of three main modules: feature extraction, feature fusion, and feature reconstruction. Specifically, the feature extraction module incorporates SPP (Spatial Pyramid Pooling) to effectively handle images of various scales. Experimental results demonstrate that the proposed model not only effectively fuses a single All-in-Focus image from images with multi focus images but also offers more efficient and robust focus fusion compared to existing methods.