• ETRI Journal
• /
• 제44권6호
• /
• pp.1004-1019
• /
• 2022

For autonomous driving without high-definition maps, we present a model capable of generating multiple plausible paths from egocentric images for autonomous vehicles. Our generative model comprises two neural networks: feature extraction network (FEN) and path generation network (PGN). The FEN extracts meaningful features from an egocentric image, whereas the PGN generates multiple paths from the features, given a driving intention and speed. To ensure that the paths generated are plausible and consistent with the intention, we introduce an attentive discriminator and train it with the PGN under a generative adversarial network framework. Furthermore, we devise an interaction model between the positions in the paths and the intentions hidden in the positions and design a novel PGN architecture that reflects the interaction model for improving the accuracy and diversity of the generated paths. Finally, we introduce ETRIDriving, a dataset for autonomous driving, in which the recorded sensor data are labeled with discrete high-level driving actions, and demonstrate the state-of-the-art performance of the proposed model on ETRIDriving in terms of accuracy and diversity.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권11호
• /
• pp.5594-5615
• /
• 2019

Malicious social robots, which are disseminators of malicious information on social networks, seriously affect information security and network environments. The detection of malicious social robots is a hot topic and a significant concern for researchers. A method based on classification has been widely used for social robot detection. However, this method of classification is limited by an unbalanced data set in which legitimate, negative samples outnumber malicious robots (positive samples), which leads to unsatisfactory detection results. This paper proposes the use of generative adversarial networks (GANs) to extend the unbalanced data sets before training classifiers to improve the detection of social robots. Five popular oversampling algorithms were compared in the experiments, and the effects of imbalance degree and the expansion ratio of the original data on oversampling were studied. The experimental results showed that the proposed method achieved better detection performance compared with other algorithms in terms of the F1 measure. The GAN method also performed well when the imbalance degree was smaller than 15%.

• ETRI Journal
• /
• 제41권4호
• /
• pp.415-425
• /
• 2019

Generative adversarial networks (GAN) have been successfully used in many image restoration tasks, including image denoising, super-resolution, and compression artifact reduction. By fully exploiting its characteristics, state-of-the-art image restoration techniques can be used to generate images with photorealistic details. However, there are many applications that require faithful rather than visually appealing image reconstruction, such as medical imaging, surveillance, and video coding. We found that previous GAN-training methods that used a loss function in the form of a weighted sum of fidelity and adversarial loss fails to reduce fidelity loss. This results in non-negligible degradation of the objective image quality, including peak signal-to-noise ratio. Our approach is to alternate between fidelity and adversarial loss in a way that the minimization of adversarial loss does not deteriorate the fidelity. Experimental results on compression-artifact reduction and super-resolution tasks show that the proposed method can perform faithful and photorealistic image restoration.

• 한국산업융합학회 논문집
• /
• 제26권1호
• /
• pp.35-46
• /
• 2023

A model test is performed to measure the sloshing impact pressure in the liquid tank. A preprocessing is performed to learn the model test results applied with various environmental conditions. In this study, we propose a method for generating data similar to the total pressure data using Generative Adversarial Networks. In addition, after approximating the generated result to the three parameter Weibull distribution, the difference of the three parameters was compared through the RMSE and SMAPE calculation results. As a result, the distribution of the generated data showed similar results to the total pressure data distribution.

• 한국전자통신학회논문지
• /
• 제18권2호
• /
• pp.349-354
• /
• 2023

최근 인테리어에 관심을 가지는 인구가 증가함에 따라 세계적으로 인테리어 시장이 크게 성장하고 있으며, 글로벌 인테리어 업체들은 다양한 인테리어 요소에 대한 시뮬레이션 서비스를 개발하여 제공하고 있다. 벽지의 디자인은 가장 중요한 인테리어 요소임에도 불구하고, 기존 벽지 디자인 시뮬레이션 서비스들은 예상되는 결과물과 실제 결과물 간 차이, 긴 시뮬레이션 작업시간, 전문적인 기술의 필요 등의 단점으로 인해 사용에 어려움이 있다. 본 논문에서는 벽지 인테리어 시뮬레이션을 위한 Cycle GAN(: Generative Adversarial Networks) 기반의 벽지 이미지 변환 기법을 제안한다. 제안하는 기법은 다양한 모양의 벽지가 사용된 인테리어 이미지 데이터를 기반으로 모델을 학습하여, 사용자에게 짧은 시간 내에 벽지 인테리어 시뮬레이션을 제공할 수 있다.

• 국제초고층학회논문집
• /
• 제12권3호
• /
• pp.203-208
• /
• 2023

The implementation of artificial intelligence (AI) design for tall building structures is an essential solution for addressing critical challenges in the current structural design industry. Generative AI technology is a crucial technical aid because it can acquire knowledge of design principles from multiple sources, such as architectural and structural design data, empirical knowledge, and mechanical principles. This paper presents a set of AI design techniques for building structures based on two types of generative AI: generative adversarial networks and graph neural networks. Specifically, these techniques effectively master the design of vertical and horizontal component layouts as well as the cross-sectional size of components in reinforced concrete shear walls and frame structures of tall buildings. Consequently, these approaches enable the development of high-quality and high-efficiency AI designs for building structures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권7호
• /
• pp.2547-2567
• /
• 2021

It has been widely acknowledged that occlusion impairments adversely distress many face recognition algorithms' performance. Therefore, it is crucial to solving the problem of face image occlusion in face recognition. To solve the image occlusion problem in face recognition, this paper aims to automatically de-occlude the human face majority or discriminative regions to improve face recognition performance. To achieve this, we decompose the generative process into two key stages and employ a separate generative adversarial network (GAN)-based network in both stages. The first stage generates an initial coarse face image without an occlusion mask. The second stage refines the result from the first stage by forcing it closer to real face images or ground truth. To increase the performance and minimize the artifacts in the generated result, a new refine loss (e.g., reconstruction loss, perceptual loss, and adversarial loss) is used to determine all differences between the generated de-occluded face image and ground truth. Furthermore, we build occluded face images and corresponding occlusion-free face images dataset. We trained our model on this new dataset and later tested it on real-world face images. The experiment results (qualitative and quantitative) and the comparative study confirm the robustness and effectiveness of the proposed work in removing challenging occlusion masks with various structures, sizes, shapes, types, and positions.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.23-30
• /
• 2022

Glaucoma is a chronic neuropathy that affects the optic nerve which can lead to blindness. The detection and prediction of glaucoma become possible using deep neural networks. However, the detection performance relies on the availability of a large number of data. Therefore, we propose different frameworks, including a hybrid of a generative adversarial network and a convolutional neural network to automate and increase the performance of glaucoma detection. The proposed frameworks are evaluated using five public glaucoma datasets. The framework which uses a Deconvolutional Generative Adversarial Network (DCGAN) and a DenseNet pre-trained model achieves 99.6%, 99.08%, 99.4%, 98.69%, and 92.95% of classification accuracy on RIMONE, Drishti-GS, ACRIMA, ORIGA-light, and HRF datasets respectively. Based on the experimental results and evaluation, the proposed framework closely competes with the state-of-the-art methods using the five public glaucoma datasets without requiring any manually preprocessing step.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2020년도 제61차 동계학술대회논문집 28권1호
• /
• pp.11-14
• /
• 2020

The importance of data in the development of deep learning is very high. Data with high morphological features are usually utilized in the domains where careful lens calibrations are needed by a human to capture those data. Synthesis of high morphological data for that domain can be a great asset to improve the classification accuracy of systems in the field. Unsupervised learning can be employed for this task. Generating photo-realistic objects of interest has been massively studied after Generative Adversarial Network (GAN) was introduced. In this paper, we propose Morpho-GAN, a method that unifies several GAN techniques to generate quality data of high morphology. Our method introduces a new suitable training objective in the discriminator of GAN to synthesize images that follow the distribution of the original dataset. The results demonstrate that the proposed method can generate plausible data as good as other modern baseline models while taking a less complex during training.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2017년도 하계학술대회
• /
• pp.36-39
• /
• 2017

2014 년 Ian Goodfellow 가 발표한 한편의 논문은 머신러닝 분야에 새로운 방향을 제시하였다. Generative Adversarial Networks, 일명 GAN 이라 불리는 이 논문은 이전까지 딥러닝으로 하지못했던 새로운 것을 창조해내는 작업을 하는 첫번째 딥러닝 알고리즘이다. 이전까지는 딥러닝을 통해 영상에서 객체의 종류를 판단하는 Classification 문제나, 영상에서 특정 객체를 검출하여 위치를 찾는 Object detection, 영상 내 특정 객체만 분리해내는 Image segmentation 문제를 해결하고 있었다. GAN 의 등장으로, 다양한 방면에서 GAN 을 적용하여 기존에는 하지 못했던 새로운 분야에 딥러닝을 적용한 사례들이 등장하고 있다. 본 논문에서는 GAN 의 원리 분석과 GAN 을 응용하여 여러 분야에 적용한 사례들을 살펴보고자 한다.