• Review of KIISC
• /
• v.30 no.5
• /
• pp.79-92
• /
• 2020

최근 전 세계적으로 '가짜뉴스', '가짜 연예인 음란 동영상' 및 '지인 능욕'에 사용되는 인공지능 기반의 딥페이크(Deepfakes)기술이 사회적인 이슈로 대두되고 있다. 딥페이크 기술이란 딥러닝 기술을 이용해 악의적으로 조작된 음성, 영상, 이미지 등을 만들어 내는 방법으로, 인공지능 기술의 발전에 맞추어 더욱더 빠르고 정교한 생성 기술이 등장하고 있다. 이러한 딥페이크 기술은 빠른 개발 속도와 쉬운 접근성을 기반으로 다양한 범죄에 악용되고 있다. 본 논문에서는 다양한 딥페이크 생성 기술을 설명하고, 이를 효율적으로 탐지 할 수 있는 다양한 데이터 기반 딥페이크 탐지 기술의 현황을 설명한다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.173-178
• /
• 2022

Recently, various computing technologies such as artificial intelligence, big data, and IoT are developing. In particular, artificial intelligence-based deepfake technology is being used in various fields such as the content and medical industry. Deepfake technology is a combination of deep learning and fake, and is a technology that synthesizes a person's face or body through deep learning, which is a core technology of AI, to imitate accents and voices. This paper uses deepfake technology to study the creation of virtual characters through the synthesis of animation models and real person photos. Through this, it is possible to minimize various cost losses occurring in the animation production process and support writers' work. In addition, as deepfake open source spreads on the Internet, many problems emerge, and crimes that abuse deepfake technology are prevalent. Through this study, we propose a new perspective on this technology by applying the deepfake technology to children's material rather than adult material.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.560-563
• /
• 2022

인공지능 기반의 딥페이크(Deepfakes) 기술이 사회적인 이슈로 대두되고 있다. 하지만 기존 딥페이크 탐지기는 sharpening, additive noise와 같은 간단한 이미지 변형만으로 탐지 우회가 가능한 문제점이 있다. 본 논문에서는 안티 포렌식에 강인한 딥페이크 탐지기를 개발하기 위해 이미지 편집 도구 기반의 안티 포렌식 데이터셋을 생성하고 적대적 학습을 수행하는 방법을 제안한다. 실험 결과를 통해 안티 포렌식에 취약한 기존 딥페이크 탐지기 성능이 제안한 적대적 학습 기법을 수행한 이후에 탐지율이 크게 개선된 것을 확인할 수 있었다.

• The Journal of the Korea Contents Association
• /
• v.22 no.8
• /
• pp.14-32
• /
• 2022

This study aims to examine the perception of false information and deepfakes according to the experience of being exposed to false information and deepfake images for YouTube content users. The study used the data from 'YouTube Use and False Information Exposure Experience' conducted by the Korea Press Foundation in 2018. For the statistical analysis, correspondent analysis was employed. The main results followed as: First, it was found that men who have been exposed to false information are most seriously aware of the problems caused by false information on YouTube. Second, regarding the need for regulation on deepfake images, women who have experienced exposure to deepfake images tended to agree, and women had a stronger awareness of the need for regulation due to damage to deepfake images than men. While YouTube users generally agree that regulation is necessary, it is required to educate YouTube users about the types of disinformation and deepfakes. In particular, it is considered to be desirable to create an environment for the self-regulation of the producers and distributors.

• Journal of The Korean Association of Information Education
• /
• v.25 no.2
• /
• pp.405-412
• /
• 2021

To prevent problems such as portrait rights, copyright, and cyber violence, an ethics education STEAM projects using deepfake apps using AI technology were developed and applied. The Deepfake apps were screened, and the contents of the elementary school curriculum were reconstructed. The STEAM project as creative experiential activities was mainly operated by the UCC activities, and applied the info-ethics awareness measurement test based on the planned behavior theory. The social STEAM project as money (financial) education was qualitatively analyzed. It was found that this STEAM classes using AI technology app significantly enhances the ethical awareness of information communication.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.724-726
• /
• 2022

딥페이크(deepfake)로 인한 디지털 범죄는 날로 교묘해지면서 사회적으로 큰 파장을 불러일으키고 있다. 이때, 딥러닝 기반 모델의 오류를 발생시키는 적대적 공격(adversarial attack)의 등장으로 딥페이크를 탐지하는 모델의 취약성이 증가하고 있고, 이는 매우 치명적인 결과를 초래한다. 본 연구에서는 2 가지 방법을 통해 적대적 공격에도 영향을 받지 않는 강인한(robust) 모델을 구축하는 것을 목표로 한다. 모델 강화 기법인 적대적 학습(adversarial training)과 영상처리 기반 방어 기법인 크기 변환(resizing), JPEG 압축을 통해 적대적 공격에 대한 강인성을 입증한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.652-654
• /
• 2022

소셜 미디어와 스마트폰의 대중화로 인해 디지털 이미지와 비디오를 만들어 내는 일이 매우 흔해졌다. 전통적인 이미지 포렌식 기술 압축 방법은 데이터를 손상시킨다는 점에서 비디오에 적용하기 부적절하다. 따라서 본 논문에서는 딥러닝과 MesoNet을 이용한 모델을 통해 참 혹은 거짓만 나타내는 기존의 결과 산출 방법에서 더 나아가 네가지의 분류 방법으로 딥페이크 감지 흐름을 살펴보고자 한다.

• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.13-19
• /
• 2023

In this paper, we investigated the implementation of a hairstyle recommendation system based on big data and deepfake technology. The proposed hairstyle recommendation system recognizes the facial shapes based on the user's photo (image). Facial shapes are classified into oval, round, and square shapes, and hairstyles that suit each facial shape are synthesized using deepfake technology and provided as videos. Hairstyles are recommended based on big data by applying the latest trends and styles that suit the facial shape. With the image segmentation map and the Motion Supervised Co-Part Segmentation algorithm, it is possible to synthesize elements between images belonging to the same category (such as hair, face, etc.). Next, the synthesized image with the hairstyle and a pre-defined video are applied to the Motion Representations for Articulated Animation algorithm to generate a video animation. The proposed system is expected to be used in various aspects of the beauty industry, including virtual fitting and other related areas. In future research, we plan to study the development of a smart mirror that recommends hairstyles and incorporates features such as Internet of Things (IoT) functionality.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.465-466
• /
• 2023

AI 기술의 발전으로 AI가 그린 그림과 인간이 직접 그린 그림을 식별하는 것이 어려워졌다. AI 기술을 통해 작품을 특정 화풍으로 그리는 것이 쉬워져 작품 도용과 평가 절하가 증가하고 있으며, AI가 인간과 유사하게 그림을 표현하는 경우 딥페이크 피싱과 같은 악용 사례도 늘어나고 있다. 따라서 본 논문에서는 AI 그림을 식별하기 위한 인공지능 모델 개발을 목표로 하고 있으며, 데이터셋을 구축하여 인공지능 기술을 활용한 알고리즘을 개발한다. YOLO Segmentation과 CNN을 활용하여 학습을 진행하고, 이를 통해 도용과 딥페이크 피해를 방지하는 프로세스를 제안한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1053-1065
• /
• 2020

With the recent development of hardware performance and artificial intelligence technology, sophisticated fake videos that are difficult to distinguish with the human's eye are increasing. Face synthesis technology using artificial intelligence is called Deepfake, and anyone with a little programming skill and deep learning knowledge can produce sophisticated fake videos using Deepfake. A number of indiscriminate fake videos has been increased significantly, which may lead to problems such as privacy violations, fake news and fraud. Therefore, it is necessary to detect fake video clips that cannot be discriminated by a human eyes. Thus, in this paper, we propose a deep-fake detection model applied with Bidirectional Convolution LSTM and Attention Module. Unlike LSTM, which considers only the forward sequential procedure, the model proposed in this paper uses the reverse order procedure. The Attention Module is used with a Convolutional neural network model to use the characteristics of each frame for extraction. Experiments have shown that the model proposed has 93.5% accuracy and AUC is up to 50% higher than the results of pre-existing studies.