• 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.

• The Journal of the Korea Contents Association
• /
• v.16 no.5
• /
• pp.35-45
• /
• 2016

The purpose of this study is to examine factors that influence mobile video services' satisfaction. 362 users who use both DMB and mobile video services complete the questionnaires measuring 6 attribute level satisfactions and total satisfaction of mobile video services, TV contents broadcasting status in mobile video services. The data are analyzed by using frequency analysis, Chi-square test, t-test/F-test, multiple regression and moderated multiple regression. The results of the study are as follows: First, diverse contents, image quality, video streaming quality, user convenience, and data consumption volume affect total satisfaction of mobile video services. Second, TV contents broadcasting status in mobile video services doesn't show positive moderation effect on the relationship between attribute level satisfaction and total satisfaction. Based on the results, the implications and limitations of this study are discussed.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.801-803
• /
• 2003

In this paper, we present a method for generating and providing spatially encoded video data that can be effectively used by GIS applications. We collect the video data by a mobile mapping system called 4S-Van that is equipped by GPS, INS, CCD camera, and DVR system. The information about spatial object appearing in video, such as occupied region in each frame, attribute value, and geo-coordinate, are generated and encoded. We suggest methods that can generate such data for each frame in semi-automatic manner. We adopt standard MPEG-7 metadata format for representation of the spatially encoded video data to be generally used by GIS application. The spatial and attribute information encoded to each video frame can make visual browsing between map and video possible. The generated video data can be provided and applied to various GIS applications where location and visual data are both important.

• Journal of Korean Society for Geospatial Information Science
• /
• v.6 no.2 s.12
• /
• pp.11-20
• /
• 1998

Although aerial photogrammetry has been used to generate or update digital maps. It is difficult to make the spatial and attribute data for all kinds of objects on the ground with only aerial photogrammetry. Therefore, we are getting informations of the object on the ground through an on-the-spot survey In order to improve accuracy and reliability of on-the-spot survey in this study, we obtained stereo images from high resolution digital camera (1152*864 pixels) and developed the video photogrammetry which was able to determine the three dimensional coordinates from stereo images by applying DLT(Direct Linear Transformation). Also, the developed video photogrammetry could generate and update the spatial and attribute data in digital maps by using a function that could connect three dimensional coordinates with the attribute data.

• Journal of the Korean Society of Costume
• /
• v.54 no.1
• /
• pp.53-68
• /
• 2004

The purposes of this study were to classify the contents of lifestyle of adolescence consumers, to examine the differences of lifestyle, importance of store attribute, preference of fashion store according to demographic variables, to test preference of fashion store and lifestyle, preference of fashion store, and to examine how the preference of fashion store was influenced by lifestyle, importance of store attribute, and demographic variables. Subjects were 317 adolescence consumers(13-24 years) in Seoul, Korea. Preference of imported & brand goods, orientation of fashion & beauty life, and family orientation had a direct effect on preference of department store($R^2$=.129). Sex had an indirect effect on preference of department store through orientation of fashion & beauty life. Search of video media & information and age(-) had a direct effect on preference of entertainment shopping center($R^2$=.150). Orientation of fashion & beauty life(-), sales promotion, sex, and reasonable consumption had a direct effect on preference of discount store(R =$^2$.109). Sex(-) had an indirect discount store through orientation of fashion & beauty life, and sex(-), search of video media & information, age(-), and reasonable consumption did through sales promotion. Sales promotion, preference of imported & brand goods(-), and store atmosphere(-) had a direct effect on preference of traditional market($R^2$=.060). Sex(-), age(-), search of video media & information, and reasonable consumption had an indirect effect on preference of traditional market through sales promotion, and search of video media & information, reasonal consumption, and acceptance of foreign culture did through store atmosphere.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1996.06b
• /
• pp.77-81
• /
• 1996

In order to conform to the needs of effective program production in multimedia era, we are studying Desk Top Program Production system. With the DTPP, users can easily produce multimedia program including video, sound, and ancillary data, and freely handle video images synthesizing video components retrieved from video database. This paper describes the new program production system, DTPP and its key technologies such as cooperative program production via multimedia network, indexing and utilization of attribute information of images, and image segmentation and spatio-temporal editing.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.2
• /
• pp.137-143
• /
• 2020

The purpose of this study is to analyze the structural relationship among advertising attribute, advertising attitude, product attitude and purchase intention in YouTube advertising. The survey subjects to achieve the purpose of this study were selected the group of 300 students who have been watching YouTube VOD once more and are attending in three universities that are located in Seoul, Gyeonggi and Chungnam. Data processing was done with SPSS 23 for frequency analysis, Cronbach's α analysis. Also, AMOS 18 was used for confirmatory factor analysis and structural equation model analysis. information, irrtation and individuality affect viewers' advertisement attitude, their product attitude and purchasing intent, among advertisement attributes of e-sports items perceived by viewers of YouTube's e-sports videos with the insertion of advertisements that cannot be skipped in the middle of the video.Therefore, the similarity between the video and the advertised product or brand should be taken into account when producing advertisements inserted in the YouTube video, and a more positive advertising effect can be achieved when advertisement can be carried out reflecting the characteristics of the target viewers.

• The Journal of Society for e-Business Studies
• /
• v.12 no.4
• /
• pp.85-98
• /
• 2007

This study uses conjoint analysis technique to calculate relative importance of each attribute of a specific online content: video-on-demand (VoD) services of TV soap operas. A series of interviews were conducted to identify and derive critical attributes for VoD services. Conjoint instrument was constructed using attributes identified and administered among actual viewers. Price, genre, additional service, and actors were found to be the most highly valued attributes of the VoD services. While price was the most valued attribute for female viewers, it was genre for male viewers. The result suggests that these attribute differentiation would lead to more profitable pricing scheme, and development of additional services may be critical for online content business' success.

• The KIPS Transactions:PartD
• /
• v.14D no.7
• /
• pp.727-732
• /
• 2007

Most of previous work on indexing and searching time series focused on the similarity matching and retrieval of one-attribute time series. However, multimedia databases such as music, video need to handle the similarity search in multi-attribute time series. The limitation of the current similarity models for multi-attribute sequences is that there is no consideration for attributes' sequences. The multi-attribute sequences are composed of several attributes' sequences. Since the users may want to find the similar patterns considering attributes's sequences, it is more appropriate to consider the similarity between two multi-attribute sequences in the viewpoint of attributes' sequences. In this paper, we propose the similarity search method based on attributes's sequences in multi-attribute time series databases. The proposed method can efficiently reduce the search space and guarantees no false dismissals. In addition, we give preliminary experimental results to show the effectiveness of the proposed method.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.151-156
• /
• 2002

Recently, as the geographic information system (GIS) which searches, manages geographic information is used more widely, there is more requests for some systems which can search and display more actual and realistic information. As a response to these requests, the video geographic information system which connects video data obtained by using cameras and geographic information as it is by displaying the obtained video data is being more popular. However, because most existing video geographic information systems consider video data as an attribute of geographic information or use simple one-way links from geographic information to video data to connect video data with geographic information, they support only displaying video data through searching geographic information. In this paper, we design and implement a video geographic information system which connects video data with geographic information and supports hi-directional search; searching geographic information through searching video data and searching video data through searching geographic information. To do this, we 1) propose an ER data model to represent connection information related to video data, geographic information, 2) propose a process to extract and to construct connection information from video data and geographic information, 3) show a component based system architecture to organize the video geographic information system.