• 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.

• 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 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 Digital Convergence
• /
• v.19 no.5
• /
• pp.229-239
• /
• 2021

For this study, we have extensively analyzed the presentation data of the government's financial-related departments and the data of each financial institution and electronic financial institution.. As a result, In Korea's electronic financial environment, real changes such as first) expansion of non-face-to-face finance, second) teleworking in the financial sector, third) abolition of accredited certification, fourth) advanced voice phishing, fifth) openness of the financial industry and diversification of forms, sixth) the'walletless society'. In addition to the above, however, global changes triggered by the Fourth Industrial Revolution spread to the financial security sector, making it difficult to respond to problems such as artificial intelligence/ deep learning/ user analysis/ deepfake technology. As the proportion of electronic finance is increasing socially, it should be studied in the fields of electronic finance and its environment, and crime and criminal investigation.

• Journal of Multimedia Information System
• /
• v.8 no.2
• /
• pp.85-92
• /
• 2021

In a face, there is much information of person's identity. Because of this property, various tasks such as expression recognition, identity recognition and deepfake have been actively conducted. Most of them use the exact frontal view of the given face. However, various directions of the face can be observed rather than the exact frontal image in real situation. The profile (side view) lacks information when comparing with the frontal view image. Therefore, if we can generate the frontal face from other directions, we can obtain more information on the given face. In this paper, we propose a combined style model based the conditional generative adversarial network (cGAN) for generating the frontal face from multi-view images that consist of characteristics that not only includes the style around the face (hair and beard) but also detailed areas (eye, nose, and mouth).

• 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.

• International journal of advanced smart convergence
• /
• v.13 no.1
• /
• pp.23-36
• /
• 2024

Interest in virtual humans continues to increase due to the development of generative AI, extended reality, computer graphics technology, and the spread of a converged metaverse that goes beyond the boundaries between reality and virtuality. Despite the negative public opinion that virtual humans were just temporary form of entertainment event in the early days of their emergence, the reason they are showing continuous growth is due to the unique characteristics of virtual humans and the expansion of diverse usage from technological advancements. The production of video content using virtual humans is becoming vigorously active, but currently there is limitation and no exact process for the technology to apply virtual humans to video content for it to be produced accordingly to the characteristics or situations of virtual humans. In this study, we investigated the characteristics of virtual human production technology methods & processes, and identifying the impact of each production technology on the production environment through examples of virtual human content applied to domestic and international video contents. In conclusion, by proposing an appropriate production method for each content, we hope to develop and assist production practitioners so they can effectively use virtual humans in video content production.

• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.535-546
• /
• 2019

Recently, there have been a lot of researches on the whole face replacement system, but it is not easy to obtain stable results due to various attitudes, angles and facial diversity. To produce a natural synthesis result when replacing the face shown in the video image, technologies such as face area detection, feature extraction, face alignment, face area segmentation, 3D attitude adjustment and facial transposition should all operate at a precise level. And each technology must be able to be interdependently combined. The results of our analysis show that the difficulty of implementing the technology and contribution to the system in facial replacement technology has increased in facial feature point extraction and facial alignment technology. On the other hand, the difficulty of the facial transposition technique and the three-dimensional posture adjustment technique were low, but showed the need for development. In this paper, we propose four facial replacement models such as 2-D Faceswap, OpenPose, Deekfake, and Cycle GAN, which are suitable for the Cox platform. These models have the following features; i.e. these models include a suitable model for front face pose image conversion, face pose image with active body movement, and face movement with right and left side by 15 degrees, Generative Adversarial Network.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.1-9
• /
• 2021

Recently, new types of security crimes have been steadily occurring in Korea due to environmental changes such as the development of IT technology, and responding to these threats has become a key national task for the establishment of a safe society as well as individuals and businesses. Meanwhile, as the Zero-Contact Era has arrived since the COVID-19 Pandemics, a new convergence security threat that combines the characteristics of the Zero-Contact Era with the existing IT development is threatening our society. Research to prevent and correct these new levels of threats is continuously required in this study. Chapter 1 examined the causes of new convergence security threats and Chapter 2 discussed cybersecurity, fake news, remote voting·work and video security threats as five major threats.

• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.248-252
• /
• 2022

The environment refers to what is surrounded by something during human life. This environment is related to the way humans live, and presents various problems on how to perceive the surrounding environment and how the behaviors that constitute the environment support the elements necessary for human life. Humans have an interest in the supportability of the environment as the interrelationship increases as humans perceive and understand the environment and accept the factors supported by the environment. In space, human movement starts from one space to the next and exchanges stimuli and reactions with the environment until reaching a target point. These human movements start with subjective judgment and during gait movement, the spatial environment surrounding humans becomes a collection of information necessary for humans and gives stimulation. will do. In this process, in particular, humans move along the movement path through movement in space and go through displacement perception and psychological changes, and recognize a series of spatial continuity. An image of thinking is formed[1]. In this process, spatial experience is perceived through the process of filtering by the senses in the real space, and the result of cognition is added through the process of subjective change accompanied by memory and knowledge, resulting in human movement. As such, the spatial search behavior begins with a series of perceptual and cognitive behaviors that arise in the process of human beings trying to read meaning from objects in the environment. Here, cognition includes the psychological process of sorting out and judging what the information is in the process of reading the meaning of the external environment, conditions, and material composition, and perception is the process of accepting information as the first step. It can be said to be the cognitive ability to read the meaning of the environment given to humans. Therefore, if we can grasp the perception of space while moving and human behavior as a response to perception, it will be possible to predict how to grasp it from a human point of view in a space that does not exist. Modern people have the theme of reminiscing dog-friendly hotels for the healing of petloss syndrome, and this thesis attempts to approach the life of companions.