• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권8호
• /
• pp.4026-4040
• /
• 2019

In this paper, we present an approach for detecting slide transitions in lecture videos by introducing the spatio-temporal residual networks. Given a lecture video which records the digital slides, the speaker, and the audience by multiple cameras, our goal is to find keyframes where slide content changes. Since temporal dependency among video frames is important for detecting slide changes, 3D Convolutional Networks has been regarded as an efficient approach to learn the spatio-temporal features in videos. However, 3D ConvNet will cost much training time and need lots of memory. Hence, we utilize ResNet to ease the training of network, which is easy to optimize. Consequently, we present a novel ConvNet architecture based on 3D ConvNet and ResNet for slide transition detection in lecture videos. Experimental results show that the proposed novel ConvNet architecture achieves the better accuracy than other slide progression detection approaches.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.88-90
• /
• 2021

본 논문은 교육 영상의 자막을 한글 형태소 분석기를 통해 추출하고 추출된 형태소 정보를 바탕으로 영상의 카테고리를 분류하는 방안에 대해 소개한다. 시스템에서 사람의 실수로 잘못된 정보가 입력되어 아이템의 특성으로 반영하게 될 경우 추천 시스템에서 정확도의 문제를 미치는 경우들이 있다. 이를 방지하기 위해 미리 분류된 영상에서 추출한 형태소 정보를 이용하여 각 카테고리에 해당하는 키워드 테이블을 생성하고, 각 카테고리 키워드 테이블과 영상의 형태소의 유사도를 비교하여 가장 유사도가 높은 키워드 테이블을 이용해 교육 영상의 카테고리를 분류한다. 이를 통해서 사람의 개입을 줄이고 시스템이 직접 영상을 분류하여 추천 시스템의 정확도를 높이는 것을 목표로 한다.

• International Journal of Computer Science & Network Security
• /
• 제21권8호
• /
• pp.87-96
• /
• 2021

The demand for e-learning through video lectures is rapidly increasing due to its diverse advantages over the traditional learning methods. This led to massive volumes of web-based lecture videos. Indexing and retrieval of a lecture video or a lecture video topic has thus proved to be an exceptionally challenging problem. Many techniques listed by literature were either visual or audio based, but not both. Since the effects of both the visual and audio components are equally important for the content-based indexing and retrieval, the current work is focused on both these components. A framework for automatic topic-based indexing and search depending on the innate content of the lecture videos is presented. The text from the slides is extracted using the proposed Merged Bounding Box (MBB) text detector. The audio component text extraction is done using Google Speech Recognition (GSR) technology. This hybrid approach generates the indexing keywords from the merged transcripts of both the video and audio component extractors. The search within the indexed documents is optimized based on the Naïve Bayes (NB) Classification and K-Means Clustering models. This optimized search retrieves results by searching only the relevant document cluster in the predefined categories and not the whole lecture video corpus. The work is carried out on the dataset generated by assigning categories to the lecture video transcripts gathered from e-learning portals. The performance of search is assessed based on the accuracy and time taken. Further the improved accuracy of the proposed indexing technique is compared with the accepted chain indexing technique.

• 공학교육연구
• /
• 제25권5호
• /
• pp.94-103
• /
• 2022

In this study, we examined the cases of online video classes conducted under the COVID-19 situation for 46 first-year students majoring in electrical engineering at a 3-year college in Gyeongnam. The research was carried out according to the course development process so that classes could be conducted efficiently, focusing on electromagnetism(1) course in the second semester of the first year. The online class consisted of uploading PPT recorded videos and lecture materials to the LMS and giving assignments. The contents of the study were in the scope of examining learners' perceptions through online learning contents, assignment submission, mutual feedback between instructors and learners, university-level lecture evaluation, and free opinions on online video-based electromagnetism(1) classes. As a result of the study, it was confirmed that online lecture review and problem-solving are particularly important for understanding lecture materials, and also given the nature of engineering subjects that require plenty of mathematical expressions, the detailed and immediate feedback provided by instructor enhances learners' class satisfaction and understanding. Based on these, the direction of online classes to be developed after the COVID-19 pandemic and teaching and learning methods suitable for the characteristics of subjects were discussed.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2007년도 춘계종합학술대회
• /
• pp.619-622
• /
• 2007

PDA는 비교적 작은 크기로 인하여 휴대하기 쉽고 간편하다는 장점이 있어 학습자에게 언제, 어디서나 학습할 수 있는 기회를 제공한다. 그러나 작은 스크린 사이즈로 인하여 동영상에서 제공되는 학습 내용을 정확하게 인식하기에는 많은 제약이 있다. 본 논문에서는 이러한 PDA의 문제점을 해결하기 위해 관심 영역(ROI)의 추출(Detection)과 이미지 스케일링(Image Scaling) 기법을 이용하여 학습에 필요한 영역을 중심으로 강의 영상의 크기를 적응적으로 변환하고 전송하는 원격 강의 시스템을 제안한다. 제안된 시스템은 802.11b 무선 네트워크상에서 실험을 통해 기존의 방법에서 보다 최적화된 강의 영상을 제공할 수 있음을 증명한다.

• 컴퓨터교육학회논문지
• /
• 제16권3호
• /
• pp.23-30
• /
• 2013

인터넷 확산으로 기존 윤리와는 다른 인터넷윤리가 하나의 학문으로 인식되고 있다. 초중고를 포함하여 대학에서도 인터넷윤리가 교양과목으로 자리잡고 있으며, 본 연구자도 이 과목을 수년째 강의하고 있다. 컴퓨터과학자로서 인터넷윤리를 학생들에게 효과적으로 전달하기 위하여, 인터넷윤리 관련 동영상을 활용하고 일부 시범 실습을 실시하는 수업을 진행하였다. 최신 UCC 제시 및 이에 대한 토론도 진행하여 학생들의 인터넷윤리 이슈와 문제점을 인식시켰다. 수강생 대상의 설문조사를 통하여, 제시된 수업방식이 학생들의 관심 유도에 효과적이며 미래 인터넷윤리 실천의지 향상에 도움이 됨을 확인하였다. 본 연구는 인터넷윤리 주제들을 대학생들에게 효과적으로 전달하는 수업 사례를 제시했다는 의의를 가진다.

• International Journal of Computer Science & Network Security
• /
• 제22권3호
• /
• pp.141-146
• /
• 2022

In modern educational practices, the issue of dependence on the experience of using multimedia by students and the adoption of technologies in education, the perception of their benefits and effectiveness in blended learning is little covered. The purpose of the academic paper lies in assessing the audiovisual context of multimedia technologies, its acceptance by students in practice on the example of using video lectures in blended learning. The methodology is based on an online survey of 120 students of Ukrainian universities who have assessed the experience level in using video lectures, as well as the constructs as follows: Technology Characteristics, Fit, Perceived Usefulness, Perceived Ease of Use, Attitude, Intention to Use, Actual Use. The results show that the majority of students use video lectures to a certain extent in their training (20,8% have used technology to a certain extent, 49,2% have often used technology in training, 20% are regular users of technology). It has been revealed that most students agree with the relevance of video lectures, the accuracy of lectures, the brevity of lectures, the clarity of lectures, as well as the high quality of lecture videos. It has been estimated that 42,5% believe that lecture videos are an effective tool towards supporting students in hybrid learning. 26,7% of students consider video lectures to be appropriate technologies for online / hybrid courses. In general, 37,5% of respondents find video lectures useful; however, 35,0% do not agree with this statement. 83,3% of students have rated the high level of ease of access to video. In total, 95% of students find lecture videos easy to use. In general, positive attitude of students to video lectures has been revealed.

• 인터넷정보학회논문지
• /
• 제19권5호
• /
• pp.77-87
• /
• 2018

Since the inception of the platform business in educational contents, the prominence of the online educational video has flipped the educational environment. Educational contents have been produced on the internet and allowed learners to access more flexible and student-centered. In fact, the number of people watching the educational content online, such as TED talks and YouTube, has increased during the past decade. The ways of delivering the lecture and the course information in online educational videos are totally different from the traditional lectures. In this paper, we aimed to examine and categorize the online educational videos based on the user's engagement and interest in the course contents. For the study, a negative binomial regression analysis was applied to estimate the effects of the attributes of the traditional lectures by comparatively analyzing the educational videos online. Several values are determined as engaging factors in the online educational videos; hybrid production of education and entertainment, shorter duration, and the number of presenters. From the study, we suggests how to produce engaging educational contents which will appeal the attentions from the users. Moreover, the result of the study may use as a guide to the providers making the productive educational videos.

• 복식문화연구
• /
• 제27권6호
• /
• pp.599-614
• /
• 2019

The aim of this study to provide basic reference data for the development of video contents used in pattern drafting education and to explore the possibility of utilizing YouTube videos in such education. Subject videos were selected using the number of views. A total of 596 videos and 28 channels were analyzed for the period July to September 2019 and the results are as follows. With regard to content, there were 27 pattern drafting items, the majority being dress, pants, skirt, blouse and sleeve drafting, although high-level content such as cowl, bustier, corset patterns were also available. Therefore, there is a high likelihood that YouTube videos could be used as educational material, especially as supplementary references to provide specific examples and easy explanations for difficult concepts or method, for students majoring in this field. However, as most videos currently focus on a few items, expanding video content to features a wider variety of clothing items at different levels is necessary. With regard to video length, it mostly ranged from 10 to 15 minutes. It is not advisable to create lengthy lecture-style videos expounding on different principles or variations in pattern drafting when developing educational video material.

• 한국전자통신학회논문지
• /
• 제16권1호
• /
• pp.81-88
• /
• 2021

최근 COVID-19로 인한 온라인 강의 영상과 같은 많은 영상이 생성되고 있는데 노동 시간의 한계와 비용의 부족 등으로 인해 자막을 보유한 영상이 일부분에 불과하여 청각장애인들의 정보 취득에 방해 요소로 대두되고 있다. 본 논문에서는 음성인식을 이용하여 자막을 자동으로 생성하고 종결 어미와 시간을 이용해 문장을 분리하여 자막을 생성함으로써 자막 생성에 드는 시간과 노동력을 줄일 수 있도록 하는 시스템을 개발하고자 한다.