Gye-Young Kim

Prof. Dr. Gye-Young Kim


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 January 2016 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
SKSul-Ho Kim
Sul-Ho Kim
KAKwon-Jae An
Kwon-Jae An
SJSeok-Woo Jang
Seok-Woo Jang
GKGye-Young Kim
Gye-Young Kim

16 June 2020 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
SLSamuel Lee
Samuel Lee
SJSeok-Woo Jang
Seok-Woo Jang
DKDongho Kim
Dongho Kim
HHHernsoo Hahn
Hernsoo Hahn
GKGye-Young Kim
Gye-Young Kim

21 July 2017 in International Journal of Distributed Sensor Networks
Verified
Journal Article
Open access
SJSeok-Woo Jang
Seok-Woo Jang
GKGye-Young Kim
Gye-Young Kim
This article proposes an intelligent monitoring system for semiconductor manufacturing equipment, which determines spec-in or spec-out for a wafer in process, using Internet of Things–based big data analysis. The proposed system consists of three phases: initialization, learning, and prediction in real time. The initialization sets the weights and the effective steps for all parameters of equipment to be monitored. The learning performs a clustering to assign similar patterns to the same class. The patterns consist of a multiple time-series produced by semiconductor manufacturing equipment and an after clean inspection measured by the corresponding tester. We modify the Line, Buzo, and Gray algorithm for classifying the time-series patterns. The modified Line, Buzo, and Gray algorithm outputs a reference model for every cluster. The prediction compares a time-series entered in real time with the reference model using statistical dynamic time warping to find the best matched pattern and then calculates a predicted after clean inspection by combining the measured after clean inspection, the dissimilarity, and the weights. Finally, it determines spec-in or spec-out for the wafer. We will present experimental results that show how the proposed system is applied on the data acquired from semiconductor etching equipment.

18 March 2025 in Sensors
Verified
Journal Article
Open access
PDF file
SJSeok-Woo Jang
Seok-Woo Jang
LYLimin Yan
Limin Yan
GKGye-Young Kim
Gye-Young Kim
In this study, we propose a dynamic scene deblurring approach using a deep supervised attention network. While existing deep learning-based deblurring methods have significantly outperformed traditional techniques, several challenges remain: (1) Invariant weights: Small conventional neural network (CNN) models struggle to address the spatially variant nature of dynamic scene deblurring, making it difficult to capture the necessary information. A more effective architecture is needed to better extract valuable features. (2) Limitations of standard datasets: Current datasets often suffer from low data volume, unclear ground truth (GT) images, and a single blur scale, which hinders performance. To address these challenges, we propose a multi-scale, end-to-end recurrent network that utilizes supervised attention to recover sharp images. The supervised attention mechanism focuses the model on features most relevant to ambiguous information as data are passed between networks at difference scales. Additionally, we introduce new loss functions to overcome the limitations of the peak signal-to-noise ratio (PSNR) estimation metric. By incorporating a fast Fourier transform (FFT), our method maps features into frequency space, aiding in the recovery of lost high-frequency details. Experimental results demonstrate that our model outperforms previous methods in both quantitative and qualitative evaluations, producing higher-quality deblurring results.

11 September 2024 in IEEE Access
Verified
Journal Article
Open access
JLJinyoung Lee
Jinyoung Lee
GKGye-Young Kim
Gye-Young Kim
Video prediction presents a formidable challenge, requiring effectively processing spatial and temporal information embedded in videos. While recurrent neural networks (RNN) and transformer-based models have been extensively explored to address spatial changes over time, recent advancements in convolutional neural networks (CNNs) have yielded high-performance video prediction models. CNN-based models offer advantages over RNN and transformer-based models due to their ease of parallel processing and lower computational complexity, highlighting their significance in practical applications. However, existing CNN-based video prediction models typically treat the spatiotemporal channels of videos similarly to the channel axis of static images. They stack frames in temporal order to construct a spatiotemporal axis and employ standard 1x1 convolution operations. Nevertheless, this approach has its limitations. Applying 1x1 convolution directly to the spatiotemporal axis results in a mixture of temporal and spatial information, which may lead to computational inefficiencies and reduced accuracy. Additionally, this operation needs to improve in processing temporal data. This study introduces a CNN-based time series decomposition model for video prediction. The proposed model first divides the 1x1 convolution operation within the channel aggregation module to independently process the temporal and spatial dimensions. To capture evolving features, the temporal axis is segregated into trend and residual components, followed by applying a time series decomposition forecasting method. To assess the performance of the proposed technique, experiments were conducted using the moving MNIST, KTH, and KITTI-Caltech benchmark datasets. In the experiments on moving MNIST, despite a reduction of approximately 55% in the number of parameters and 37% in computational cost, the proposed method improved accuracy by up to 7% compared to the previous approach.

14 November 2019 in The Journal of Supercomputing
Verified
Journal Article
Closed access
PDF file
SKSul-Ho Kim
Sul-Ho Kim
SJSeok-Woo Jang
Seok-Woo Jang
JPJin-Ho Park
Jin-Ho Park
GKGye-Young Kim
Gye-Young Kim