• ETRI Journal
• /
• v.42 no.1
• /
• pp.67-77
• /
• 2020

Image captioning has received significant interest in recent years, and notable results have been achieved. Most previous approaches have focused on generating visual descriptions from images, whereas a few approaches have exploited visual descriptions for image classification. This study demonstrates that a good performance can be achieved for both description generation and image classification through an end-to-end joint learning approach with a loss function, which encourages each task to reach a consensus. When given images and visual descriptions, the proposed model learns a multimodal intermediate embedding, which can represent both the textual and visual characteristics of an object. The performance can be improved for both tasks by sharing the multimodal embedding. Through a novel loss function based on class activation mapping, which localizes the discriminative image region of a model, we achieve a higher score when the captioning and classification model reaches a consensus on the key parts of the object. Using the proposed model, we established a substantially improved performance for each task on the UCSD Birds and Oxford Flowers datasets.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.17 no.4
• /
• pp.501-509
• /
• 2014

In NCW(Network Centric Warfare) environment, mobile nodes communicate through wireless link. But wireless link provides limited networking performance due to signal interferences or mobility of nodes. So it is quite challenge to acquire enough networking resources and use the resources efficiently. In this paper, we have proposed a P2P based tactical information sharing system which provides satisfactory visual information playout for mobile nodes(i.e., military personnel, vehicle,..) in NCW environment. Our proposed system consists of two components. One is caching-enabled switch which stores tactical information segments at its internal storage and then transports them to mobile nodes when require. Another is centralized scheduling algorithm which exploits networking resources more efficiently. To validate performance of proposed system, we performed series of experiments in wireless network testbed. Results show improved performance in terms of segment-missing ratio, networking resources usage, sharing time, and number of simultaneous playout mobile nodes with acceptable playout continuity(i.e., over 95%).

• Asia pacific journal of information systems
• /
• v.25 no.1
• /
• pp.119-141
• /
• 2015

With the rapid spread of social media, video-sharing social media like YouTube has emerged as a consumption and distribution channel for entertainment goods such as music videos and movie trailers. In tourism research, there has been a lot of research of how the visual media such as movies and soap operas induced tourism. However, no studies have attempted to examine the role of social media as a music consumption channel and its impact on tourism. Expanding a body of media-induced tourism, we analysed the impact of video-sharing social media on music-induced tourism with a case of Korean pop music and inbound tourism to Korea. Developing a Web-crawler, we collected YouTube users' comments data on 166 Korean pop music video clips which were released from 2009 to 2012 with over 1 million view counts. Controlling many of the determinants of tourism and analysing country-by-country impact of YouTube comments with the panel data, we found that engagement of Korean pop music video clips on YouTube is a significant predictor for the flow of inbound tourists to Korea.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1061-1066
• /
• 2004

It is certain that the future manufacturing environment will be network-centric and spatially distributed based on Internet. Today, wide variety of distributed computing and communication technologies are available for implementing a system for product data exchange and sharing. One of the technologies that have been received most attentions for product data exchange and sharing is Product Data Management (PDM). PDM tries to integrate and manage process of data and technical documents that are connected to physical product components. In accordance to previous researches about PDM, it can be regarded as an integration tool of many different areas, which ensures that the right information is available to the right person at the right time and in the right form throughout the enterprise. PDM with Web-enabled CAD system is proposed in this paper in order to acknowledge the usefulness of the system mentioned. The system will use Web service on Visual Studio C#.Net to invoke the web application system.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2004.05a
• /
• pp.245-248
• /
• 2004

It is certain that the future manufacturing environment will be network-centric and spatially distributed based on Internet. Today, wide variety of distributed computing and communication technologies are available for implementing a system for product data exchange and sharing. One of the technologies that have been received most attentions for product data exchange and sharing is Product Data Management (PDM). PDM tries to integrate and manage process of data and technical documents that are connected to physical product components. In accordance to previous researches about PDM, it can be regarded as an integration tool of many different areas, which ensures that the right information is available to the right person at the right time and in the right form throughout the enterprise. PDM with Web-enabled CAD system is proposed in this paper in order to acknowledge the usefulness of the system mentioned. The system will use Web service on Visual Studio C#.Net to invoke the web application system.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.2181-2182
• /
• 2011

This paper deals with power integration of large urban railway systems for improvement of regenerative energy utilization. The main idea of this paper is that the integration of railway power systems will increase efficiency through sharing regenerative energy. To verify the effects of this system, case studies for Seoul Metro Line has been performed. Where railway data of Seoul Subway Line 5 and Line 7 are used. In this work, the integrated railway system is modelled using Compaq Visual Fortran.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3357-3376
• /
• 2015

Wireless sensor network is an important research topic that has attracted a lot of attention in recent years. However, most of the interest has focused on wireless sensor network to gather scalar data such as temperature, humidity and vibration. Scalar data are insufficient for diverse applications such as video surveillance, target recognition and traffic monitoring. However, if we use camera sensors in wireless sensor network to collect video data which are vast in information, they can provide important visual information. Video sensor networks continue to gain interest due to their ability to collect video information for a wide range of applications in the past few years. However, how to efficiently store the massive data that reflect environmental state of different times in video sensor network and how to quickly search interested information from them are challenging issues in current research, especially when the sensor network environment is complicated. Therefore, in this paper, we propose a fast algorithm for extracting key frames from video and describe the design and implementation of key frame extraction and sharing in Android for wireless video sensor network.

• The Journal of the Korea Contents Association
• /
• v.19 no.5
• /
• pp.82-93
• /
• 2019

Fashion cannot encourage co-consumption with other people as a personal item, but it can lead to new consumer needs if fashion sharing service can professionally replace the time and cost of purchasing and managing goods. The purpose of this study is to empirically investigate the factors influencing the acceptance of fashion-sharing services based on the integration theory of user acceptance and utilization (UTAUT), and to discuss the virtuous cycle and sustainability pursuit of resources through the activation of the sharing. In this study, the research model for the acceptance of fashion sharing applications is schematized, and the survey was conducted 300 women aged 20~49 years. The screens of 'Project Anne', a representative fashion sharing service in Korea, were provided as a visual data. Reliability analysis, correlation analysis, confirmatory factor analysis, structural equation analysis, and multiple group analysis were performed using SPSS 23.0 and AMOS 22.0 statistical package for statistical analysis. As a result, efficiency and social influence positively influenced behavioral intention to use, and age has found that efficiency and social influences modulate the intensity of behavioral intention to use. Therefore, for the consumer acceptance and activation of fashion sharing services, marketing activities emphasizing efficiency and strengthening social influence factors are essential. Also, it is necessary to maintain the existing target group, 30~40s, and also construct additional products and price services for the 20s. This study is of academic significance in presenting basic data for empirical research on consumer acceptance of fashion sharing, and suggests a study on the influence relationship among user experience components for real users in the future.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.6
• /
• pp.1546-1553
• /
• 1998

In this paper, we investigate a human identification scheme using visual cryptography, which can decode concealed images without and crytographic computations. T.Katoh & H.Imai proposed a secret sharing scheme which can conceal two query images into one displayed image for the identification. the generalized construction of the share matrix is proposed to conceal the multiple query images by stacking user slides o the displayed image. Furthermore, an extended methodis proposed in which group of slides can conceal an independent secret image as well as the pairs of slides can.

• ETRI Journal
• /
• v.34 no.3
• /
• pp.466-469
• /
• 2012

In this letter, we propose a dual autostereoscopic display platform employing a natural interaction method, which will be useful for sharing visual data with users. To provide 3D visualization of a model to users who collaborate with each other, a beamsplitter is used with a pair of autostereoscopic displays, providing a visual illusion of a floating 3D image. To interact with the virtual object, we track the user's hands with a depth camera. The gesture recognition technique we use operates without any initialization process, such as specific poses or gestures, and supports several commands to control virtual objects by gesture recognition. Experiment results show that our system performs well in visualizing 3D models in real-time and handling them under unconstrained conditions, such as complicated backgrounds or a user wearing short sleeves.