• Broadcasting and Media Magazine
• /
• v.28 no.1
• /
• pp.53-60
• /
• 2023

최근 영상처리 및 컴퓨터비전 등 많은 분야에서 딥러닝 기술이 빠르게 발전하면서 다양한 문제들을 높은 성능으로 해결하고 있다. 이에 MPEG (Moving Picture Experts Group) 표준에서도 딥러닝 기반의 미디어 기술이 활발히 제안 및 논의되고 있다. 특히, 몰입형 입체영상 압축을 위한 MPEG-I (MPEG Immersive) 표준은 메타버스 산업으로 크게 관심받고 있는 가상현실, 증강현실, 그리고 혼합현실 등에 대응하기 위해 현재 활발히 연구 중이다. 입체영상은 일반적으로 복수 시점의 컬러영상과 깊이영상으로 구성되어 있어 데이터의 양이 크기 때문에, MPEG-I 표준은 시점 간의 중복된 영역들을 제거하는 프루닝 과정을 통해 효율적인 압축을 수행한다. 하지만, 프루닝 과정에서 정반사 영역이 함께 제거되는 문제로 정확한 입체영상 복원에 한계가 있다. 본 학회지에서는 이러한 문제점을 해결하기 위하여 MPEG-I 표준에 기고된 딥러닝 기반의 정반사 영역 검출을 통한 몰입형 입체영상 압축에 대해 소개한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.369-372
• /
• 2023

본 논문에서는 삼각형 구조로 구성된 3차원 메쉬(Mesh)에서 합성곱 신경망(Convolution Neural Network, CNN)을 응용하여 정확도가 높은 새로운 학습 표현 기법을 제시한다. 우리는 메쉬를 구성하고 있는 폴리곤의 edge와 face의 로컬 특징을 기반으로 학습을 진행한다. 일반적으로 딥러닝은 인공신경망을 수많은 계층 형태로 연결한 기법을 말하며, 주요 처리 대상은 1, 2차원 데이터 형태인 오디오 파일과 이미지였다. 인공지능에 대한 연구가 지속되면서 3차원 딥러닝이 도입되었지만, 기존의 학습과는 달리 3차원 딥러닝은 데이터의 확보가 쉽지 않다. 혼합현실과 메타버스 시장의 확대로 인해 3차원 모델링 시장이 증가하고, 기술의 발전으로 데이터를 획득할 수 있는 방법이 생겼지만, 3차원 데이터를 직접적으로 학습에 이용하는 방식으로 적용하는 것은 쉽지 않다. 그렇게 때문에 본 논문에서는 산업 현장에서 이용되는 데이터인 메쉬 구조를 폴리곤의 최소 단위인 삼각형 형태로 구성하여 학습 데이터를 구성해 기존의 방법보다 정확도가 높은 학습 기법을 제안한다.

• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.115-135
• /
• 2023

The purpose of this study is to assess the effectiveness of using deep learning language models to extract references automatically and create a reference database for research reports in an efficient manner. Unlike academic journals, research reports present difficulties in automatically extracting references due to variations in formatting across institutions. In this study, we addressed this issue by introducing the task of separating references from non-reference phrases, in addition to the commonly used metadata extraction task for reference extraction. The study employed datasets that included various types of references, such as those from research reports of a particular institution, academic journals, and a combination of academic journal references and non-reference texts. Two deep learning language models, namely RoBERTa+CRF and ChatGPT, were compared to evaluate their performance in automatic extraction. They were used to extract metadata, categorize data types, and separate original text. The research findings showed that the deep learning language models were highly effective, achieving maximum F1-scores of 95.41% for metadata extraction and 98.91% for categorization of data types and separation of the original text. These results provide valuable insights into the use of deep learning language models and different types of datasets for constructing reference databases for research reports including both reference and non-reference texts.

• Journal of The Korean Association of Information Education
• /
• v.26 no.6
• /
• pp.533-541
• /
• 2022

Metaverse have begun to attract attention because it facilitate interaction between learners and teachers in non face- to- face environment. In order to use the metaverse in the educational field such as online class, it is important that pre-service teachers intend to use it. The purpose of this study is to analyze the structural relationship between the pre-service teacher's educational competence and the intention to use the metaverse based on the technology acceptance model. The influence factors of flexibility for new technology, teacher efficacy, and TPACK were examined. It was conducted with 240 pre-service teachers, and the data of 183 pre-service teachers finally collected were used for the analysis. As a result of the study, among the metaverse educational competencies of elementary school pre-service teachers, flexibility and TPACK mediate perceived ease, and the pathways affecting metaverse use intention were significant. In this regard, theoretical and practical implications that can be helpful in the discussion and intention of using the metaverse of pre-service teachers were presented.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.245-248
• /
• 2012

Called G-learning the domestic history of the serious game is only a mere 10-odd years or so but in recent years, due to the increased interest of users and industry for the serious game and serious games industry has been the continued growth of large quantity. Current the various types of serious games have been developed and the demand for serious games is increasing rapidly. Therefore that is becoming an important issue the problem of including a quality evaluation elements into serious game metadata to make the right choice of users for the high-quality serious games. In this paper, we recognize deeply the need for quality evaluation element development for serious game and we propose a method and a process for developing the quality evaluation elements of serious game. Also, we develop a draft of the quality evaluation elements for serious game using proposed method.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.24-31
• /
• 2023

Meta-learning is metacognition that instantly distinguishes between knowing and unknown. It is a learning method that adapts and solves new problems by self-learning with a small amount of data.A few-shot learning method is a type of meta-learning method that accurately predicts query data even with a very small support set. In this study, we propose a method to solve the limitations of the prototype created with the mean-point vector of each class. For this purpose, we use the few-shot learning method that created the prototype used in the few-shot learning method as the median prototype. For quantitative evaluation, a handwriting recognition dataset and mini-Imagenet dataset were used and compared with the existing method. Through the experimental results, it was confirmed that the performance was improved compared to the existing method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.188-190
• /
• 2022

We propose how to configure the AI education system based on the purpose of hybrid learning and the teaching-learning principle. Based on the four components of hybrid learning, we have designed the system conceptual diagram and DB configuration diagram for on-line and offline learning environments for effective AI education. The proposed AI education system model in this paper is expected to be a foundation for maximizing the effectiveness of AI education according to the level and needs of learners and building a more effective learner-centered learning environment in cultivating computational thinking in AI education.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.212-215
• /
• 2021

본 논문에서는 딥러닝의 CNN(Convolution Neural Network) 학습을 통하여 악성코드를 실행시키지 않고서 악성코드 변종을 패밀리 그룹으로 분류하는 방법을 연구한다. 먼저 데이터 전처리를 통해 3가지의 서로 다른 방법으로 악성코드 이미지와 메타데이터를 생성하고 이를 CNN으로 학습시킨다. 첫째, 악성코드의 byte 파일을 8비트 gray-scale 이미지로 시각화하는 방법이다. 둘째, 악성코드 asm 파일의 opcode sequence 정보를 추출하고 이를 이미지로 변환하는 방법이다. 셋째, 악성코드 이미지와 메타데이터를 결합하여 분류에 적용하는 방법이다. 이미지 특징 추출을 위해서는 본고에서 제안한 CNN을 통한 학습 방식과 더불어 3개의 Pre-trained된 CNN 모델을 (InceptionV3, Densnet, Resnet-50) 사용하여 전이학습을 진행한다. 전이학습 시에는 마지막 분류 레이어층에서 본 논문에서 선택한 데이터셋에 대해서만 학습하도록 파인튜닝하였다. 결과적으로 가공된 악성코드 데이터를 적용하여 9개의 악성코드 패밀리로 분류하고 예측 정확도를 측정해 비교 분석한다.

• Journal of Advanced Navigation Technology
• /
• v.13 no.5
• /
• pp.781-786
• /
• 2009

The cloud computing environment is a new trend of web based application parts. It can be IT business model that is able to easily support learning service and allocate resources through the internet to users. U-learning also is a maximal model with efficiency of the internet based learning. Thus, in this research, we proposed a design of u-learning's teaching and learning model that is applying the internet based learning. Proposal method is to fit u-learning and has 7 steps: Preparing, planning, gathering, learning process, analysis and evaluation, and feedback. We make a cloud u-learning server and cloud LMS to process and manage the service. And We also make a mobile devices meta data to aware the model.

• Journal of Digital Contents Society
• /
• v.7 no.2
• /
• pp.125-131
• /
• 2006

In the recent year, the need for effective searching for e-learning contents has emerged as a revitalization of e-learning. Therefore, there are many studies for the contents searching method that make use of semantic web technology. Also, there are many studies to apply the ontology as the core technology of sematic web to e-learning. In this study, we bind the KEM(Korea Education Metadata) that enacted as the first technical standard of e-learning, KS X7001, in Dec. 2004 with OWL to use the metadata for a effective searching with ontology through the solving of the problem of metadata. And, we discuss some problems that occur in progress of binding and the solutions.