• 전기전자학회논문지
• /
• 제27권3호
• /
• pp.313-318
• /
• 2023

본 논문은 Vision Transformer를 기반으로 하는 Video Classification의 성능을 개선하는 방법으로 fine-tuning를 적용한 신경망을 제안한다. 최근 딥러닝 기반 실시간 비디오 영상 분석의 필요성이 대두되고 있다. Image Classification에 사용되는 기존 CNN 모델의 특징상 연속된 Frame에 대한 연관성을 분석하기 어렵다는 단점이 있다. 이와 같은 문제를 Attention 메커니즘이 적용된 Vistion Transformer와 Non-local 신경망 모델을 비교 분석하여 최적의 모델을 찾아 해결하고자 한다. 또한, 전이 학습 방법으로 fine-tuning의 다양한 방법을 적용하여 최적의 fine-tuning 신경망 모델을 제안한다. 실험은 UCF101 데이터셋으로 모델을 학습시킨 후, UTA-RLDD 데이터셋에 전이 학습 방법을 적용하여 모델의 성능을 검증하였다.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제11권3호
• /
• pp.177-184
• /
• 2007

The purpose of this paper is to introduce the development of a new Lesson Video-Case Lab (LVCL) program in Suzhou, China. This program involves the creation of several series of hypermedia video-cases on teaching and learning designed to facilitate mathematics teachers' professional development. Each of these video-cases consists of lesson clips, case questions, interviews with experts, comments by peers, responses by students and other related resources. The study has implications pertaining to the use of technology in teacher development, the production of hypermedia video-cases, as well as research on case-based pedagogy and pedagogy in general.

• 복식문화연구
• /
• 제18권1호
• /
• pp.161-176
• /
• 2010

This study attempts to develop e-learning contents for the "fashion information analysis," which should be performed in combination of theory and practice in fashion education, in a recent Web-based education environment. The e-learning contents are developed by using the three steps of analysis, design and development. In order to analyze the requirements of curricula, the status and content of fashion information related curricular provided by general universities and cyber universities are analyzed, and then on the basis of this analysis the education objectives and detailed contents of e-learning contents are determined. Design and development of e-learning contents consist of the types, education menus and education form. The types of e-learning contents are developed into an education form combined with tutorial and practice/activity types. The education menus are comprised of syllabic, the profiles of professors and students, announcements, questions and answers regarding respect subjects, materials, and notes. Meanwhile, the education form consists of the following sections: "Before the Learning", "Learning", "Review Questions", "Final Review", "References", and "Practice", "Before the Learning" consists of two parts of a theory lecture that is composed of video clips and Power-Point presentations, and practice that consists of an audio(or video) lecture and illustration or Photoshop software execution window.

• 한국간호교육학회지
• /
• 제19권1호
• /
• pp.43-51
• /
• 2013

Purpose: The purpose of this study was to examine the effects of peer feedback on nursing students using videos filmed during the catheterization procedure. Method: The study was a structured nonequivalent group, pretest-posttest design. The experimental group was filmed performing foley catheterization. After the procedure, the group video review followed and peer feedback was provided. Results: The experimental group had a higher mean score in learning motivation than the control group. However, the differences in learning attitude and self efficacy were not statistically significant. Conclusion: The results of this study indicate that video-aided peer feedback was effective in enhancing students' motivation to learn psychomotor skills.

• 대한공업교육학회지
• /
• 제30권2호
• /
• pp.104-114
• /
• 2005

The purposes of this study were to examin instructor and students's responses and to explore the pros and cons and the possibility of the DVC(desktop video conferencing) instruction at a C university in Daejon. 6 males and 5 females took the course. 10 lived in Daejeon and 1, who accessed from other region, lived other province, Pusan. Students were surveyed and interviewed, instructor was interviewed and the online observation of the instruction was undertaken. Students' responses showed that the DVC instruction was very effective to generate student's learning motivation and interest and to deliver the learning contents. Instructor's variety roles different from conventional one were very important and instruction must be run to activate students' interactions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제15권4호
• /
• pp.341-355
• /
• 2011

A research group consisting of some mathematics education scholars and school teachers has been formed to build a website which intended to combine both theories and practices of mathematical teaching since 2002. There were three working stages: video-tapping realistic math teaching, developing video discs of different themes, and designing e-Learning. The group members learned knowledge of teaching by actual participating. On the other hand, the products enabled the audience to get professional development in knowledge of both mathematics and teaching. Management process of this group and effects upon the participants and users will be presented and discussed in this paper. A research group consisting of some mathematics education scholars and school teachers has been formed to build a website which intended to combine both theories and practices of mathematical teaching since 2002. There were three working stages: video-tapping realistic math teaching, developing video discs of different themes, and designing e-Learning. The group members learned knowledge of teaching by actual participating. On the other hand, the products enabled the audience to get professional development in knowledge of both mathematics and teaching. Management process of this group and effects upon the participants and users will be presented and discussed in this paper.

• 한국컴퓨터정보학회논문지
• /
• 제10권6호
• /
• pp.137-152
• /
• 2005

본 논문에서는 사용자 프로파일을 기반으로 한 정보 필터링을 사용하여 개개인의 학생들에게 맞는 효율적인 웹 기반 비디오 학습 평가 시스템을 제시한다. 비디오 기반 문제들이 주어지면, 키 프레임들은 위치, 크기, 그리고 컬러 정보를 기반으로 추출되고, 문제 출제 구간은 시간 윈도우와 그레이-레벨 히스토그램의 차이에 의해 추출된다 게다가, 카테고리-기반 시스템과 키워드 기반 시스템을 합성하여 문제 출제가 효율적인 평가를 확실해 하도록 주어진다. 따라서 학생들은 관심 영역을 유지하면서 부족한 영역을 보충함으로서 학업 성취도를 향상시킬 수 있다.

• 한국수산과학회지
• /
• 제54권6호
• /
• pp.965-973
• /
• 2021

The behavior of angelfish Pterophyllum scalare exposed to low and high temperatures was monitored by video tracking, and information such as the initial speed, changes in speed, and locations of the fish in the tank were analyzed. The water temperature was raised from 26℃ to 36℃ or lowered from 26℃ to 16℃ for 4 h. The control group was maintained at 26℃ for 8 h. The experiment was repeated five times for each group. Machine learning analysis comprising a long short-term memory model was used to train and test the behavioral data (80 s) after pre-processing. Results showed that when the water temperature changed to 36℃ or 16℃, the average speed, changes in speed and fractal dimension value were significantly lower than those in the control group. Machine learning analysis revealed that the accuracy of 80-s video footage data was 87.4%. The machine learning used in this study could distinguish between the optimal temperature group and changing temperature groups with specificity and sensitivity percentages of 86.9% and 87.4%, respectively. Therefore, video tracking technology can be used to effectively analyze fish behavior. In addition, it can be used as an early warning system for fish health in aquariums and fish farms.

• International Journal of Computer Science & Network Security
• /
• 제22권12호
• /
• pp.13-18
• /
• 2022

In the post-information era, most of technologies have a visual part, or at least some functions related to visualization. It is also one of the popular means of presenting materials in education area. However, despite its popularity, the impact of visualization on the effectiveness of learning still remains controversial. Even more controversial is its usefulness in developing creativity, which is one of the most important skills for today's employee. The authors considered the use of visualization as a tool for the development of children's creativity on the example of learning video games, in particular, ClassCraft to distinguish features that, from the point of view of psychology, may lead to developing creativity even being not useful for educational purposes. It is concluded that video games useful for learning may have features, that are inappropriate in formal educational context, but important to develop creative thinking.