• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.480-482
• /
• 2012

하이퍼네트워크(Hypernetworks)는 하이퍼에지(hyperedge)들로 이루어진 생성 모델(generative model)로서, 주로 이산(binary) 데이터에 적용되어왔다. 본 논문에서는 이산 데이터와 실수 데이터를 모두 다룰 수 있는 새로운 하이퍼네트워크 모델을 에너지 기반 모델(energy-based model)의 형태로 제시하고, 비지도 학습(unsupervised learning) 알고리즘으로 데이터를 성공적으로 학습함을 간단한 실험을 통해 보이겠다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.792-794
• /
• 2017

현재 머신러닝(Machine Learning) 기술은 기존의 머신러닝과 조합 및 변형 되어 조금 더 발전 된 형태로 연구되어지고 있다. 따라서 수많은 알고리즘이 개발되고 있는 시점이다. 본 연구는 최근 좋은 결과로 관심을 받고있는 GAN(Generative Adversarial Net)을 중심으로 IT기술의 머신러닝과 그림을 조합하여 자동채색을 목적으로 GAN 알고리즘을 비교하고 분석하고자 한다. GAN 알고리즘들 가운데서 'Conditional GAN'과 'Wasserstein GAN'을 사용하여 자동채색을 적용시켰고, 가장 부합한 알고리즘을 찾고 성능을 비교하여 어떠한 알고리즘이 '자동채색' 목적에 더 부합한지 비교하고 판단 한다.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.82.3-82.3
• /
• 2019

In this study, we generate future full disk magnetograms in 12, 24, 36 and 48 hours advance from SDO/HMI images using deep learning. To perform this generation, we apply the convolutional generative adversarial network (cGAN) algorithm to a series of SDO/HMI magnetograms. We use SDO/HMI data from 2011 to 2016 for training four models. The models make AI-generated images for 2017 HMI data and compare them with the actual HMI magnetograms for evaluation. The AI-generated images by each model are very similar to the actual images. The average correlation coefficient between the two images for about 600 data sets are about 0.85 for four models. We are examining hundreds of active regions for more detail comparison. In the future we will use pix2pix HD and video2video translation networks for image prediction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4835-4848
• /
• 2020

The recent expansion of the digital content market is increasing the technical demand for various facial image transformations within the virtual environment. The recent image translation technology enables changes between various domains. However, current image-to-image translation techniques do not provide stable performance through unsupervised learning, especially for shape learning in the face transition field. This is because the face is a highly sensitive feature, and the quality of the resulting image is significantly affected, especially if the transitions in the eyes, nose, and mouth are not effectively performed. We herein propose a new unsupervised method that can transform an in-wild face image into another face style through radical transformation. Specifically, the proposed method applies two face-specific feature loss functions for a generative adversarial network. The proposed technique shows that stable domain conversion to other domains is possible while maintaining the image characteristics in the eyes, nose, and mouth.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.11
• /
• pp.1482-1484
• /
• 2019

We present a method for transmitting 3D object information in real time in a telepresence system. Three-dimensional object information consists of a large amount of point cloud data, which requires high performance computing power and ultra-wideband network transmission environment to process and transmit such a large amount of data in real time. In this paper, multiple users can transmit object motion and facial expression information in real time even in small network bands by using GANs (Generative Adversarial Networks), a non-supervised learning machine learning algorithm, for real-time transmission of 3D point cloud data. In particular, we propose the creation of an object similar to the original using only the feature information of 3D objects using conditional GANs.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.344-358
• /
• 2022

Text summarization is the task of producing a shorter version of a long document while accurately preserving the main contents of the original text. Abstractive summarization generates novel words and phrases using a language generation method through text transformation and prior-embedded word information. However, newly coined words or out-of-vocabulary words decrease the performance of automatic summarization because they are not pre-trained in the machine learning process. In this study, we demonstrated an improvement in summarization quality through the contextualized embedding of BERT with out-of-vocabulary masking. In addition, explicitly providing precise pointing and an optional copy instruction along with BERT embedding, we achieved an increased accuracy than the baseline model. The recall-based word-generation metric ROUGE-1 score was 55.11 and the word-order-based ROUGE-L score was 39.65.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.19-20
• /
• 2020

미디어 산업의 발전으로 스토리보드와 같은 선화 이미지의 자동채색 연구가 국내외에서 진행되고 있다. 하지만 자동채색 모델 용량에 초점을 두는 연구는 아직 진행되고 있지 않다. 기존 자동채색 연구는 모델 용량이 최소 567MB 이상으로 모델 용량이 큰 단점을 가지고 있다. 본 논문에서는 채색을 2단계로 나누는 이중 생성자 구조와 기존 U-Net을 개선한 생성자를 사용해 기존 U-Net에 비해 30%, VGG16/19를 사용한 기법과 비교해 최대 85% 작은 106MB 모델을 생성했고 FID(Fréchet Inception Distance)를 통한 이미지 평가결과 512x512px에서 153.69의 채색성능을 얻었다.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.41.1-41.1
• /
• 2021

We generate solar magnetograms and EUV images from Galileo sunspot drawings using a deep learning model based on conditional generative adversarial networks. We train the model using pairs of sunspot drawing from Mount Wilson Observatory (MWO) and their corresponding magnetogram (or UV/EUV images) from 2011 to 2015 except for every June and December by the SDO (Solar Dynamic Observatory) satellite. We evaluate the model by comparing pairs of actual magnetogram (or UV/EUV images) and the corresponding AI-generated one in June and December. Our results show that bipolar structures of the AI-generated magnetograms are consistent with those of the original ones and their unsigned magnetic fluxes (or intensities) are well consistent with those of the original ones. Applying this model to the Galileo sunspot drawings in 1612, we generate HMI-like magnetograms and AIA-like EUV images of the sunspots. We hope that the EUV intensities can be used for estimating solar EUV irradiance at long-term historical times.

• Journal of the Korean Society of Systems Engineering
• /
• v.19 no.2
• /
• pp.145-156
• /
• 2023

Trajectory of ballistic missile is defined by inherent flight dynamics, which decided range and maneuvering characteristics. It is crucial to predict range and maneuvering characteristics of ballistic missile in KAMD (Korea Air and Missile Defense) to minimize damage due to ballistic missile attacks, Nowadays, needs for applying AI(Artificial Intelligence) technologies are increasing due to rapid developments of DNN(Deep Neural Networks) technologies. To apply these DNN technologies amount of data are required for superviesed learning, but trajectory data of ballistic missiles is limited because of security issues. Trajectory data could be considered as multivariate time series including many variables. And augmentation in time series data is a developing area of research. In this paper, we tried to augment trajectory data of ballistic missiles using recently developed methods. We used TimeVAE(Time Variational AutoEncoder) method and TimeGAN(Time Generative Adversarial Networks) to synthesize missile trajectory data. We also compare the results of two methods and analyse for future works.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.5
• /
• pp.1-9
• /
• 2019

In this paper, we propose a video enhancement method using generative adversarial networks to remove raindrops and restore the background information on the removed region in the coastal wave video imagery distorted by raindrops during rainfall. Two experimental models are implemented: Pix2Pix network widely used for image-to-image translation and Attentive GAN, which is currently performing well for raindrop removal on a single images. The models are trained with a public dataset of paired natural images with and without raindrops and the trained models are evaluated their performance of raindrop removal and background information recovery of rainwater distortion of coastal wave video imagery. In order to improve the performance, we have acquired paired video dataset with and without raindrops at the real coast and conducted transfer learning to the pre-trained models with those new dataset. The performance of fine-tuned models is improved by comparing the results from pre-trained models. The performance is evaluated using the peak signal-to-noise ratio and structural similarity index and the fine-tuned Pix2Pix network by transfer learning shows the best performance to reconstruct distorted coastal wave video imagery by raindrops.