• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1339-1342
• /
• 2022

예술적 스타일 전이는 예술 작품이 지닌 특징을 다른 이미지에 적용하는 이미지 처리의 오랜 화두 중 하나로, 최근에는 StyleGAN 과 같이 미리 학습된 GAN(생성적 적대 신경망)을 통해 제한된 데이터로도 고해상도의 예술적 초상화를 생성하도록 학습하는 연구가 다양한 방면에서 성과를 내고 있다. 본 논문에서는 2 가지 경로의 StyleGAN과 Facial Destylization 을 통해 고해상도의 예시 기반 스타일 전이를 달성한 DualStyleGAN 연구에 대해 소개하고, 기존 연구에서 사용된 Facial Destylization 방법이 지닌 한계점을 분석한 뒤, 이를 개선한 새로운 방법, Re-Destyle을 제안한다. 새로운 Re-Destyle 방법으로 Facial Destylization 을 적용할 경우 학습 시간을 기존 연구의 방법보다 20 배 이상 개선할 수 있으며 그 결과 1000 개 이하의 적은 데이터와 1~2 시간의 추가 학습만으로도 원하는 타겟 초상화 스타일에 대해 1024×1024 수준의 고해상도의 예시 기반 초상화 스타일 전이 및 이미지 생성 모델을 학습할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.380-382
• /
• 2019

GAN(Generative Adversarial Network)은 정해진 학습 데이터에서 정해진 생성자와 구분자가 서로 각각에게 적대적인 관계를 유지하며 동시에 서로에게 생산적인 관계를 유지하며 가능한 긍정적인 영향을 주며 학습하는 기계학습 분야이다. 전통적인 문장 생성은 단어의 통계적 분포를 기반으로 한 마르코프 결정 과정(Markov Decision Process)과 순환적 신경 모델(Recurrent Neural Network)을 사용하여 학습시킨다. 이러한 방법은 문장 생성과 같은 연속된 데이터를 기반으로 한 모델들의 표준 모델이 되었다. GAN은 표준모델이 존재하는 해당 분야에 새로운 모델로써 다양한 시도가 시도되고 있다. 하지만 이러한 모델의 시도에도 불구하고, 지금까지 해결하지 못하고 있는 다양한 문제점이 존재한다. 이 논문에서는 다음과 같은 두 가지 문제점에 집중하고자 한다. 첫째, Sequential 한 데이터 처리에 어려움을 겪는다. 둘째, 무작위로 생성하기 때문에 사용자가 원하는 데이터만 출력되지 않는다. 본 논문에서는 이러한 문제점을 해결하고자, 부분적인 정답 제공을 통한 조건별 생산적 적대 생성망을 설계하여 이 방법을 사용하여 해결하였다. 첫째, Sequence to Sequence 모델을 도입하여 Sequential한 데이터를 처리할 수 있도록 하여 원시적인 텍스트를 생성할 수 있게 하였다. 둘째, 부분적인 정답 제공을 통하여 문장의 생성 조건을 구분하였다. 결과적으로, 제안하는 기법들로 원시적인 감정 텍스트를 생성할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.793-794
• /
• 2019

본 시스템은 기존 안저사진의 높은 해상도와 광각 안저사진의 촬영의 편의성과 넓은 시야 각의 장점을 둘 다 취하기 위해, 광각 안저사진을 원본으로 하여 기존 안저사진을 생성하는 작업을 제안한다. 안저사진의 생성 알고리즘은 이미지 생성에 특화된 생산적 적대 신경망(GANs)을 채택하였다. 신경망의 성능을 평가하기 위해, 생성된 안저사진과 직접 촬영한 기존 안저사진으로부터 특성을 추출을 하여 특이점 간의 거리를 측정하고, 이를 바탕으로 두 사진의 유사도를 판단한다.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.9-17
• /
• 2021

With the advancement of big data analysis, artificial intelligence, machine learning, etc., data analytics technology has developed to help with optimal decision-making. However, in certain areas, the lack of data restricts the use of these techniques. For example, real estate related data often have a long release cycle because of its recent release or being a non-liquid asset. In order to overcome these limitations, we studied the scalability of the existing time series through the TimeGAN model. A total of 45 time series related to weekly real estate data were collected within the period of 2012 to 2021, and a total of 15 final time series were selected by considering the correlation between the time series. As a result of data expansion through the TimeGAN model for the 15 time series, it was found that the statistical distribution between the real data and the extended data was similar through the PCA and t-SNE visualization algorithms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.4
• /
• pp.591-597
• /
• 2022

Recently, many studies have been done on the data augmentation technique as a way to efficiently build datasets. Among them, a representative data augmentation technique is a method of utilizing Generative Adversarial Network (GAN), which generates data similar to real data by competitively learning generators and discriminators. However, when learning GAN, there are cases where a broken pixel image occurs among similar data generated according to the environment and progress, which cannot be used as a dataset and causes an increase in learning time. In this paper, an algorithm was developed to select these damaged images by analyzing the histogram of image data generated during the GAN learning process, and as a result of comparing them with the images generated in the existing GAN, the ratio of the damaged images was reduced by 33.3 times(3,330%).

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.766-769
• /
• 2021

In order to perform tasks such as design, control, optimization, and prediction of flight vehicle trajectories based on machine learning techniques including deep learning, a certain amount of flight vehicle trajectory data is required. However, there are cases in which it is difficult to secure more than a certain amount of flight vehicle trajectory data for various reasons. In such cases, synthetic data generation could be one way to make machine learning possible. In this paper, to explore this possibility, we generated and evaluated synthetic flight vehicle trajectory data using time-series generative adversarial neural network. In addition, various ablation studies (comparative experiments) were performed to explore the possibility of using synthetic data in the aircraft trajectory prediction task. The experimental results presented in this paper are expected to be of practical help to researchers who want to conduct research on the possibility of using synthetic data in the generation of synthetic flight vehicle trajectory data and the work related to flight vehicle trajectories.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.9 no.3
• /
• pp.303-310
• /
• 2021

Cracks in bridges are important factors that indicate the condition of bridges and should be monitored periodically. However, a visual inspection conducted by a human expert has problems in cost, time, and reliability. Therefore, in recent years, researches to apply a deep learning model are started to be conducted. Deep learning requires sufficient data on the situations to be predicted, but bridge crack data is relatively difficult to obtain. In particular, it is difficult to collect a large amount of crack data in a specific situation because the shape of bridge cracks may vary depending on the bridge's design, location, and construction method. This study developed a crack detection model that generates and trains insufficient crack data through a Generative Adversarial Network. GAN successfully generated data statistically similar to the given crack data, and accordingly, crack detection was possible with about 3% higher accuracy when using the generated image than when the generated image was not used. This approach is expected to effectively improve the performance of the detection model as it is applied when crack detection on bridges is required, though there is not enough data, also when there is relatively little or much data f or one class.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2022.06a
• /
• pp.383-384
• /
• 2022

Following the development of technologies such as digital twin, IoT, and AI after the 4th industrial revolution, decision-making problems are being solved based on high-dimensional data analysis. This has recently been applied to the port logistics sector, and a number of studies on big data analysis, deep learning predictions, and simulations have been conducted on container terminals to improve port productivity. These high-dimensional data analysis techniques generally require a large number of data. However, the global port environment has changed due to the COVID-19 pandemic in 2020. It is not appropriate to apply data before the COVID-19 outbreak to the current port environment, and the data after the outbreak was not sufficiently collected to apply it to data analysis such as deep learning. Therefore, this study intends to present a port data augmentation method for data analysis as one of these problem-solving methods. To this end, we generate the container stowage situation of the yard through a generative adversarial neural network model in terms of container terminal operation, and verify similarity through statistical distribution verification between real and augmented data.

• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.672-684
• /
• 2020

Restoring a low resolution and motion blurred light field has become essential due to the growing works on parallax-based image processing. These tasks are known as light-field enhancement process. Unfortunately, only a few state-of-the-art methods are introduced to solve the multiple problems jointly. In this work, we design a framework that jointly solves light field spatial super-resolution and motion deblurring tasks. Particularly, we generate a straight-forward neural network that is trained under low-resolution and 6-degree-of-freedom (6-DOF) motion-blurred light field dataset. Furthermore, we propose the strategy of local region optimization on the adversarial network to boost the performance. We evaluate our method through both quantitative and qualitative measurements and exhibit superior performance compared to the state-of-the-art methods.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.1
• /
• pp.9-18
• /
• 2021

Although Generative Adversarial Networks (GANs) have gained great popularity in computer vision and related fields, generating audio signals independently has yet to be presented. Unlike images, an audio signal is a sampled signal consisting of discrete samples, so it is not easy to learn the signals using CNN architectures, which is widely used in image generation tasks. In order to overcome this difficulty, GAN researchers proposed a strategy of applying time-frequency representations of audio to existing image-generating GANs. Following this strategy, we propose an improved method for increasing the fidelity of synthesized audio signals generated by using GANs. Our method is demonstrated on a public speech dataset, and evaluated by Fréchet Inception Distance (FID). When employing our method, the FID showed 10.504, but 11.973 as for the existing state of the art method (lower FID indicates better fidelity).