• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.7C
• /
• pp.675-686
• /
• 2005

In this paper, we propose an efficient face region detection technique for the content-based video summarization. To segment video, shot changes are detected from a video sequence and key frames are selected from the shots. We select one frame that has the least difference between neighboring frames in each shot. The proposed face detection algorithm detects face region from selected key frames. And then, we provide user with summarized frames included face region that has an important meaning in dramas or movies. Using Bayes classification rule and statistical characteristic of the skin pixels, face regions are detected in the frames. After skin detection, we adopt the projection method to segment an image(frame) into face region and non-face region. The segmented regions are candidates of the face object and they include many false detected regions. So, we design a classifier to minimize false lesion using CART. From SGLD matrices, we extract the textual feature values such as Inertial, Inverse Difference, and Correlation. As a result of our experiment, proposed face detection algorithm shows a good performance for the key frames with a complex and variant background. And our system provides key frames included the face region for user as video summarized information.

• Journal of Korea Multimedia Society
• /
• v.8 no.2
• /
• pp.269-278
• /
• 2005

In order to process video data effectively, it is required that the content information of video data is loaded in database and semantics-based retrieval method can be available for various queries of users. In this paper, we propose a video retrieval system which support semantics retrieval of various users for massive video data by user's keywords and comparison area learning based on automatic agent. By user's fundamental query and selection of image for key frame that extracted from query, the agent gives the detail shape for annotation of extracted key frame. Also, key frame selected by user becomes a query image and searches the most similar key frame through color histogram comparison and comparison area learning method that proposed. From experiment, the designed and implemented system showed high precision ratio in performance assessment more than 93 percents.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10c
• /
• pp.41-45
• /
• 2007

The rapid growing of video content production leads to the necessity of developing more complex indexing systems in order to efficiently allow searching, retrieval and presentation of the desired segments of videos. This paper presents a method for indexing soccer video through automatic extraction of information from internet. The proposed paper defines a metadata structure to formally represent the knowledge of soccer matches and provides an automatic method to extract semantic information from web-sites. This approach improves the capability to extract more reliable and richer semantic Information for soccer videos. Experimental results demonstrate that the proposed method provides an efficient performance.

• The KIPS Transactions:PartD
• /
• v.9D no.1
• /
• pp.31-42
• /
• 2002

In this paper, we propose a new spatio-temporal representation scheme which can model moving objets trajectories effectively in video data and a new signature-based access method for moving objects trajectories which can support efficient retrieval on user query based on moving objects trajectories. The proposed spatio-temporal representation scheme supports content-based retrieval based on moving objects trajectories and concept-based retrieval based on concepts(semantics) which are acquired through the location information of moving objects trajectories. Also, compared with the sequential search, our signature-based access method can improve retrieval performance by reducing a large number of disk accesses because it access disk using only retrieved candidate signatures after it first scans all signatures and performs filtering before accessing the data file. Finally, we show the experimental results that proposed scheme is superior to the Li and Shan's scheme in terns of both retrieval effectiveness and efficiency.

• International journal of advanced smart convergence
• /
• v.5 no.3
• /
• pp.61-65
• /
• 2016

In the TV broadcasting, movie production and business the transportation of video between creators (programmers, studios) and distributors (broadcast and cable networks, cable and satellites companies) is still a mix of File Transfer Protocol (FTP), physical delivery, and expensive multicast satellite. Cloud-based file sync-and-share providers like Dropbox and box are playing an increasing role, but the industry's unique demands for speed and multicasting have fueled the growth of IP Video transport. This paper gives a solid grasp of the major elements of IP video technology, including content preparation, system architecture alternatives and network performance management.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.31-32
• /
• 2022

Commercialized game engines help develop content and support various platforms, but there are performance differences depending on the function. In this paper, a VR video conferencing platform corresponding to source multi-use was implemented, and the connection delay time according to the type of game server engine and the number of platform users was measured. As a result of the measurement, the latency was the lowest at 22.146 msec when using the Amazon GameLift engine and with 2 users, and the latency was the highest at 27.007 msec when using the ProudNet engine and with 10 users.

• The KIPS Transactions:PartB
• /
• v.10B no.5
• /
• pp.543-552
• /
• 2003

An increasing in quantity of multimedia data brought a new problem that expected data should be retrieved fast and exactly. The adequate representation is a key element for the efficient retrieval. For this reason, MPEG-7 standard was established for description of multimedia data in 2001. However, the content of the standard is massive and the approach method is not clear for real application system yet, because of properties of MPEG-7 standard that has to include a lot of potential cases. In this paper, we suggested implementation scheme of retrieval system with using of only visual descriptors and presented the performance results of developed system. From the result of developed system, MPEG-7 VIRS (Video/Image Retrieval System), we analyzed the retrieval results between using individual descriptor and using multiple descriptors, and showed a layout for real application system.

• Journal of Korea Game Society
• /
• v.20 no.5
• /
• pp.13-20
• /
• 2020

Biometric technology with the advance of deep learning enabled the new types of content. Especially, face recognition can provide immersion in terms of convenience and non-compulsiveness, but most commercial content has limitations that are limited to application areas. In this paper, we attempted to overcome these limitations, implement content that can utilize face recognition technology based on realtime video feed. We used Unity engine for high quality graphics, but performance degradation and frame drop occurred. To solve them, we augmented Dlib toolkit and adjusted the resolution image.

• The KIPS Transactions:PartB
• /
• v.16B no.2
• /
• pp.101-108
• /
• 2009

As content serviced for IP set-top boxes is streamed over IP network, the existing hacking tools for IP network can be used to capture the streamed content. Until recently, most of the content serviced on IP set-top boxes has been MPEG-2 TS. However, this content will be gradually moved to WMV, MPEG-4 or H.264 because of the relatively low compression efficiency and overhead of the TS packet. In this paper, we propose a DRM scheme other than WMRM for streamed WMV content. Our approach is to design a DRM scheme independent to the existing WMV streaming system. We also design this scheme in order to provide the feature for controlling the DRM processing time considering device performance. We verified it through the experiment.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.7
• /
• pp.231-240
• /
• 2014

Video transcoding techniques provide an efficient mechanism to make a video content adaptive to the capabilities of a variety of clients. However, it is hard to provide an appropriate quality-of-service(QoS) to the clients owing to heavy workload on transcoding operations. In light of this fact, this paper presents the dual transcoding method in order to guarantee QoS in streaming services by maximizing resource usage in a transcoding server equipped with both CPU and GPU computing units. The CPU and GPU computing units have different architectural features. The proposed method speculates workload of incoming transcoding requests and then schedules the requests either to the CPU or GPU accordingly. From performance evaluation, the proposed dual transcoding method achieved a speedup of 1.84 compared with traditional transcoding approach.