• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.120-124
• /
• 2021

Recently, human attribute recognition (HAR) attracts a lot of attention due to its wide application in video surveillance systems. Recent deep-learning-based solutions for HAR require time-consuming training processes. In this paper, we propose a post-processing technique that utilizes the historical video frames to improve prediction results without invoking re-training or modifying existing deep-learning-based classifiers. Experiment results on a large-scale benchmark dataset show the effectiveness of our proposed method.

• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.428-438
• /
• 2020

The jitter in the digital videos lowers the visibility and degrades the efficiency of image processing and image compressing. In this paper, we propose a video stabilizer architecture based on triplet CNN and a method of synthesizing training datasets based on video synthesis. Compared with a conventional deep-learning video stabilization method, the proposed video stabilizer can reduce wobbling distortion.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.6
• /
• pp.1175-1181
• /
• 2004

This paper proposes a system to analyze and summarize the video shots of baseball game TV program into fifteen categories. Our System consists of three modules: feature extraction, Hidden Markov Model (HMM) training, and video shot categorization. Video Shots belongs to the same class are not necessarily similar, so we require that the training set is large enough to include video shot with all possible variations to create a robust Hidden Markov Model. In the experiments, we have illustrated that our system can recognize the 15 different shot classes with a success ratio of 84.72%.

• Proceedings of the KSR Conference
• /
• 1999.11a
• /
• pp.259-266
• /
• 1999

A study on the static type train simulator will include the training of new drives requires that the environment of the cab, controls placement, etc. must highly realistic so that driver can readily transfer his training experience to the real world. The simulator computer sends video disc speed command to a Video PC processor. A video switcher select the output of the on-line player. This selection is done with loss of vertical synchronization, meaning the picture will not noticeable roll or jump as the simulation mover from disc to disc. The video image quality remain contestant through the simulated speed range from zero to 100km/h. Flicker is avoided in the scene by the use of a TBC(Time Base Corrector) which causes the display of one video field at a time. Thus, no interfield jitter is present when the scene is stopped.

• The Journal of Korean Physical Therapy
• /
• v.32 no.2
• /
• pp.65-69
• /
• 2020

Purpose: This study examined the effects of action-observation training using the Y-Balance on the balance ability of young adults. Methods: Thirty-four healthy adults were randomized into an action-observation group (n=17) or a control group (n=17). All subjects performed the Y-Balance test before and after watching the video. The action observation group watched a video of someone performing a Y-Balance test, and the control group watched a video of scenery unrelated to the training. The subjects were measured through a Y-Balance test for both the length of the legs extended in three directions and the Y-balance composite score. Results: A significant difference in the Y-balance composite score was observed between the two groups. A part of the direction of the extended leg in the action observation group was increased significantly (posteromedial direction of the right leg, posterolateral direction of the right leg, posteromedial direction of left leg) compared to the control group. Conclusion: These results suggest that action observation training only could help improve balance.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.24 no.2
• /
• pp.128-137
• /
• 2017

Purpose: This study was conducted to investigate the effects of core basic nursing education using cellular phone video recordings on self-confidence in performance, achievement, and practice satisfaction. Methods: The research design was a nonequivalent control group modified and non-synchronized design. The participants were assigned to either the experimental (n=60) or control group (n=67). During one semester, the experimental group received training for 10 selected core basic nursing skills using cellular phone video recording. The control group, also for one semester, received training with traditional practices. Self-confidence in performance and practice satisfaction were measured using a questionnaire, and achievement was evaluated by professors at the conclusion of the semester. Collected data were analyzed using t-test and Pearson correlation coefficients. Results: Self-confidence in performance increased significantly after training in the experimental group (t=7.94, p<.001). The experimental group showed significantly higher self-confidence in performance (t=2.39, p=.018) and achievement (t=2.40, p=.018) compared to the control group. However, the difference in practice satisfaction was not statistically significant. Conclusion: The results indicate that cellular phone video recording is effective in improving self-confidence in performance and achievement of core basic nursing skills.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.161-169
• /
• 2011

Recent intelligent video surveillance system asks for development of more advanced technology for analysis and recognition of video data. Especially, machine learning algorithm such as Support Vector Machine (SVM) is used in order to recognize objects in video. Because SVM training demands massive amount of computation, parallel processing technique is necessary to reduce the execution time effectively. In this paper, we propose a parallel processing method of SVM training with a multi-core processor. The results of parallel SVM on a 4-core processor show that our proposed method can reduce the execution time of the sequential training by a factor of 2.5.

• Journal of the Korean Home Economics Association
• /
• v.39 no.4
• /
• pp.35-47
• /
• 2001

The purposes of this article were to examine teacher's guidance over video utility and to investigate problems for video use in the classroom. Two-hundred-ninety-four teachers in half-day and all-day early childhood programs participated in this study. Descriptive analysis and chi-square tests were used. Results resealed that teacher's guidance before, during, and after video viewing was not carried out enough for the effectiveness of video teaming. The major problem for using a video in early childhood programs was insufficient video materials. These data were interpreted with respect to teachers'training practices and government policy.

• Korean Journal of Child Studies
• /
• v.11 no.1
• /
• pp.87-99
• /
• 1990

The present study was designed to investigate the effect of altruistic TV viewing and verbal training on the altruistic behavior of preschoolers. The subjects of this study were a total of 56 boys and 57 girls from a kindergarten in Kwachon, Kyung-gi do. The subjects were assigned to one of three conditions: in the first condition subjects were shown video tapes designed to portray prosocial themes (TV modeling group) ; in the second condition subjects saw the video tapes in addition to verbal training (TV modeling plus verbal training group); in the third condition subjects received neither TV modeling nor verbal training (control group). Statistical analysis was by ANOVA and $Scherr\acute{e}$ test. Significant differences were found in altruistic behavior between the TV modeling and the TV modeling with verbal training groups.

• Proceedings of the Korea Database Society Conference
• /
• 1997.10a
• /
• pp.24-48
• /
• 1997

. Supported byㆍDARPA′s image Understanding (IU) program under "Video Retrieval Based on Language and image Analysis" project.DARPA′s Computer Assisted Education and Training Initiative program (CAETI)ㆍObjective: Develop practical systems for automatic understanding and indexing of video sequences using both audio and video tracks(omitted)