• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1999.11b
• /
• pp.15-20
• /
• 1999

본 논문에서는 사람의 정면 얼굴을 찍은 컬러 동영상에서 얼굴의 각 구성 요소에 대한 특징점들을 추출하여 3차원 개인 얼굴 모델을 생성하고 이를 얼굴의 표정 움직임에 따라 애니메이션 하는 방법을 제시한다. 제안된 방법은 얼굴의 정면만을 촬영하도록 고안된 헬멧형 카메라( Head-mounted camera)를 사용하여 얻은 2차원 동영상의 첫 프레임(frame)으로부터 얼굴의 특징점들을 추출하고 이들과 3차원 일반 얼굴 모델을 바탕으로 3차원 얼굴 특징점들의 좌표를 산출한다. 표정의 변화는 초기 영상의 특징점 위치와 이 후 영상들에서의 특징점 위치의 차이를 기반으로 알아낼 수 있다. 추출된 특징점 및 얼굴 움직임은 보다 다양한 응용 이 가능하도록 최근 1단계 표준이 마무리된 MPEG-4 SNHC의 FDP(Facial Definition Parameters)와FAP(Facial Animation Parameters)의 형식으로 표현되며 이를 이용하여 개인 얼굴 모델 및 애니메이션을 수행하였다. 제안된 방법은 단일 카메라로부터 촬영되는 영상을 기반으로 이루어지는 MPEG-4 기반 화상 통신이나 화상 회의 시스템 등에 유용하게 사용될 수 있다.

• Convergence Security Journal
• /
• v.24 no.3
• /
• pp.117-129
• /
• 2024

With the announcement of Transformer, a new type of architecture, in 2017, there have been many changes in language models. In particular, the development of LLM (Large language model) has enabled generative AI services such as search and chatbot to be utilized in various business areas. However, security issues such as personal information leakage and reliability issues such as hallucination, which generates false information, have raised concerns about the effectiveness of these services. In this study, we aimed to analyze the factors that are increasing the frequency of using generative AI in the workplace despite these concerns. To this end, we derived eight factors that affect the quality of LLM-based generative AI response data and empirically analyzed the impact of these factors on job satisfaction using a valid sample of 195 respondents. The results showed that expertise, accessibility, diversity, and convenience had a significant impact on intention to continue using, security, stability, and reliability had a partially significant impact, and completeness had a negative impact. The purpose of this study is to academically investigate how customer perception of response data quality affects business utilization satisfaction and to provide meaningful practical implications for customer-centered services.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.269-274
• /
• 2023

질의응답 (Question Answering)은 주어진 질문을 이해하여 그에 맞는 답변을 생성하는 자연어 처리 분야의 핵심적인 기계 독해 작업이다. 현재 대다수의 자연어 이해 작업은 사전학습 언어 모델에 미세 조정 (finetuning)하는 방식으로 학습되고, 질의응답 역시 이러한 방법으로 진행된다. 하지만 미세 조정을 통한 전이학습은 사전학습 모델의 크기가 커질수록 전이학습이 잘 이루어지지 않는다는 단점이 있다. 게다가 많은 양의 파라미터를 갱신한 후 새로운 가중치들을 저장하여야 한다는 용량의 부담이 존재한다. 본 연구는 최근 대두되는 deep prompt tuning 방법론을 한국어 추출형 질의응답에 적용하여, 미세 조정에 비해 학습시간을 단축시키고 적은 양의 파라미터를 활용하여 성능을 개선했다. 또한 한국어 추출형 질의응답에 최적의 prompt 길이를 최적화하였으며 오류 분석을 통한 정성적인 평가로 deep prompt tuning이 모델 예측에 미치는 영향을 조사하였다.

• The Korean Society of Law and Medicine
• /
• v.24 no.4
• /
• pp.67-101
• /
• 2023

The utilization of generative AI in the medical field is also being rapidly researched. Access to vast data sets reduces the time and energy spent in selecting information. However, as the effort put into content creation decreases, there is a greater likelihood of associated issues arising. For example, with generative AI, users must discern the accuracy of results themselves, as these AIs learn from data within a set period and generate outcomes. While the answers may appear plausible, their sources are often unclear, making it challenging to determine their veracity. Additionally, the possibility of presenting results from a biased or distorted perspective cannot be discounted at present on ethical grounds. Despite these concerns, the field of generative AI is continually advancing, with an increasing number of users leveraging it in various sectors, including biomedical and life sciences. This raises important legal considerations regarding who bears responsibility and to what extent for any damages caused by these high-performance AI algorithms. A general overview of issues with generative AI includes those discussed above, but another perspective arises from its fundamental nature as a large-scale language model ('LLM') AI. There is a civil law concern regarding "the memorization of training data within artificial neural networks and its subsequent reproduction". Medical data, by nature, often reflects personal characteristics of patients, potentially leading to issues such as the regeneration of personal information. The extensive application of generative AI in scenarios beyond traditional AI brings forth the possibility of legal challenges that cannot be ignored. Upon examining the technical characteristics of generative AI and focusing on legal issues, especially concerning the protection of personal information, it's evident that current laws regarding personal information protection, particularly in the context of health and medical data utilization, are inadequate. These laws provide processes for anonymizing and de-identification, specific personal information but fall short when generative AI is applied as software in medical devices. To address the functionalities of generative AI in clinical software, a reevaluation and adjustment of existing laws for the protection of personal information are imperative.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.211-216
• /
• 2024

The purpose of this study is to examine the functional aspects linked to the productivity innovation of ChatGPT, which emerged as a result of the rapid development of AI technology, and to identify ways to apply it in the jewelry industry. By analyzing the definition of ChatGPT and its features that improve productivity, I identify the scope of its application in the jewelry production process and derive meaningful implications. ChatGPT has the characteristics of 'learning', 'communication', and 'generative'. It enhances productivity by applying it to the jewelry industry. Social issues arise from the paradigm shift in the creation methods of generative AI. The version of ChatGPT is continuously upgraded along with the expansion of parameters. Accordingly, we would like to discuss ways to strengthen the competitiveness of the jewelry industry by conducting continuous research.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.13-26
• /
• 2024

Multi-modal generation is the process of generating results based on a variety of information, such as text, images, and audio. With the rapid development of AI technology, there is a growing number of multi-modal based systems that synthesize different types of data to produce results. In this paper, we present an AI system that uses speech and text recognition to describe a person and generate a montage image. While the existing montage generation technology is based on the appearance of Westerners, the montage generation system developed in this paper learns a model based on Korean facial features. Therefore, it is possible to create more accurate and effective Korean montage images based on multi-modal voice and text specific to Korean. Since the developed montage generation app can be utilized as a draft montage, it can dramatically reduce the manual labor of existing montage production personnel. For this purpose, we utilized persona-based virtual person montage data provided by the AI-Hub of the National Information Society Agency. AI-Hub is an AI integration platform aimed at providing a one-stop service by building artificial intelligence learning data necessary for the development of AI technology and services. The image generation system was implemented using VQGAN, a deep learning model used to generate high-resolution images, and the KoDALLE model, a Korean-based image generation model. It can be confirmed that the learned AI model creates a montage image of a face that is very similar to what was described using voice and text. To verify the practicality of the developed montage generation app, 10 testers used it and more than 70% responded that they were satisfied. The montage generator can be used in various fields, such as criminal detection, to describe and image facial features.

• Journal of KIISE:Databases
• /
• v.27 no.2
• /
• pp.227-237
• /
• 2000

A Web database System is a large-scaled multimedia application system that has multimedia processing facilities and cooperates with relational/Object-Oriented DBMS. Conventional hypertext modeling methods and DB gateway have limitations for Web database because of their restricted versatile presentation abilities and inefficient concurrency control caused by bottleneck in cooperation processing. Thus, we suggest a Dynamic Navigation Model & Virtual Graph Structure. The Dynamic Navigation Model supports implicit query processing and dynamic creation of navigation spaces, and introduce node-link creation rule considering navigation styles. We propose a mapping methodology between the suggested hypertext model and the relational data model, and suggest a dynamic allocation scheduling technique for query processing server based on weighted value. We show that the proposed technique enhances the retrieval performance of Web database systems in processing complex queries concurrently.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.5
• /
• pp.93-100
• /
• 2006

This paper presents a multi-sensor data simulator and a data fusion algorithm for tracking high dynamic flight target from Radar and Telemetry System. The designed simulator generates time-asynchronous multiple sensor data with different data rates and communication delays. Measurement noises are incorporated by using realistic sensor models. The proposed fusion algorithm is designed by a 21st order distributed Kalman Filter which is based on the PVA model with sensor bias states. A fault detection and correction logics are included in the algorithm for bad data and sensor faults. The designed algorithm is verified by using both simulation data and actual real data.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.3
• /
• pp.27-33
• /
• 2003

The image-based 3D modeling is the technique of generating a 3D graphic model from images acquired using cameras. It is being researched as an alternative technique for the expensive 3D scanner. In this paper, we propose the image-based, 3D modeling system using calibrated stereo cameras. The proposed algorithm for rendering, 3D model consists of three steps, camera calibration, 3D reconstruction, and 3D registration step. In the camera calibration step, we estimate the camera matrix for the image aquisition camera. In the 3D reconstruction step, we calculate 3D coordinates using triangulation from corresponding points of the stereo image. In the 3D registration step, we estimate the transformation matrix that transforms individually reconstructed 3D coordinates to the reference coordinate to render the single 3D model. As shown the result, we generated relatively accurate 3D model.

• Journal of The Korean Association For Science Education
• /
• v.36 no.3
• /
• pp.389-397
• /
• 2016

This study aims to test the efficacy of English-based automated computer scoring models and machine translation to score Korean college students' written responses on natural selection concept items. To this end, I collected 128 pre-service biology teachers' written responses on four-item instrument (total 512 written responses). The machine translation software (i.e., Google Translate) translated both original responses and spell-corrected responses. The presence/absence of five scientific ideas and three $na{\ddot{i}}ve$ ideas in both translated responses were judged by the automated computer scoring models (i.e., EvoGrader). The computer-scored results (4096 predictions) were compared with expert-scored results. The results illustrated that no significant differences in both average scores and statistical results using average scores was found between the computer-scored result and experts-scored result. The Pearson correlation coefficients of composite scores for each student between computer scoring and experts scoring were 0.848 for scientific ideas and 0.776 for $na{\ddot{i}}ve$ ideas. The inter-rater reliability indices (Cohen kappa) between computer scoring and experts scoring for linguistically simple concepts (e.g., variation, competition, and limited resources) were over 0.8. These findings reveal that the English-based automated computer scoring models and machine translation can be a promising method in scoring Korean college students' written responses on natural selection concept items.