• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.2A
• /
• pp.146-156
• /
• 2002

This paper presents an adaptive H.263+ rate control algorithm for streaming video applications under the networks supporting bandwidth renegotiation, which can communicate with end-users to accommodate their time-varying bandwidth requests during the data transmission. That is, the requests of end-users can be supported adaptively according to the availability of the network resources, and thus the overall network utilization can be improved simultaneously. They are especially suitable for the transmission of non-stationary video traffics. The proposed rate control algorithm communicates with the network to renegotiate the required bandwidth fort the underlying video which are measured based on the motion change information, and choose their control strategies according to the renegotiation results. Unlike most conventional algorithms that control only the spatial quality by adjusting quantization parameters, the proposed algorithm treats both the spatial and temporal qualities at the same time to enhance human visual perceptual quality. Experimental results are provided to demonstrate that the proposed rate control algorithm can achieve superior performance to the conventional ones with low computational complexity under the networks supporting bandwidth renegotiation.

• Journal of Information Processing Systems
• /
• v.14 no.1
• /
• pp.150-161
• /
• 2018

We propose a novel deep neural network model for detecting human activities in untrimmed videos. The process of human activity detection in a video involves two steps: a step to extract features that are effective in recognizing human activities in a long untrimmed video, followed by a step to detect human activities from those extracted features. To extract the rich features from video segments that could express unique patterns for each activity, we employ two different convolutional neural network models, C3D and I-ResNet. For detecting human activities from the sequence of extracted feature vectors, we use BLSTM, a bi-directional recurrent neural network model. By conducting experiments with ActivityNet 200, a large-scale benchmark dataset, we show the high performance of the proposed DeepAct model.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.648-651
• /
• 2023

Video anomaly detection aims to detect abnormal events. Motivated by the power of transformers recently shown in vision tasks, we propose a novel transformer-based network for video anomaly detection. To capture long-range information in video, we employ a multi-scale transformer as an encoder. A convolutional decoder is utilized to predict the future frame from the extracted multi-scale feature maps. The proposed method is evaluated on three benchmark datasets: USCD Ped2, CUHK Avenue, and ShanghaiTech. The results show that the proposed method achieves better performance compared to recent methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1730-1747
• /
• 2018

Multi-view video plus depth (MVD) is a mainstream format of 3D scene representation in free viewpoint video systems. The advanced 3D extension of the high efficiency video coding (3D-HEVC) standard introduces new prediction tools to improve the coding performance of depth video. However, the depth video in 3D-HEVC is time consuming. To reduce the complexity of the depth video inter coding, we propose a fast coding unit (CU) size and mode decision algorithm. First, an off-line trained Bayesian model is built which the feature vector contains the depth levels of the corresponding spatial, temporal, and inter-component (texture-depth) neighboring largest CUs (LCUs). Then, the model is used to predict the depth level of the current LCU, and terminate the CU recursive splitting process. Finally, the CU mode search process is early terminated by making use of the mode correlation of spatial, inter-component (texture-depth), and inter-view neighboring CUs. Compared to the 3D-HEVC reference software HTM-10.0, the proposed algorithm reduces the encoding time of depth video and the total encoding time by 65.03% and 41.04% on average, respectively, with negligible quality degradation of the synthesized virtual view.

• The KIPS Transactions:PartC
• /
• v.16C no.4
• /
• pp.487-494
• /
• 2009

VOD(Video On Demand) that provides streaming service according to the user's requirement in real time, consists of the video streaming server and the client system. Since it is very hard to apply the traditional server-client model that a server communicates with many clients through 1:1 connection to VOD system because it requires very high network bandwidth, many researches have been done to address this problem. Batching technique is one of VOD system based on Multicast that requires very small network bandwidth. However, the batching based VOD system has a limitation that it is very hard to provide VCR(Video Cassette Recorder) ability. In this paper, we propose a technique that reduces the required network bandwidth to provide VCR function by using the characteristic of MPEG, one of international video compression standard. In the proposed technique, a new video stream for VCR function is constructed with I pictures that is able to be decoded independently. The new video stream for VCR function is transmitted with the video stream for normal play together in Batching manner. The performance evaluation result shows that the proposed technique not only reduces the required network bandwidth and memory usage but also decreases the CPU usages.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.7
• /
• pp.495-502
• /
• 2013

Motion estimation (ME) that limits the performance of image quality and encoding speed has been developed to reduce temporal redundancy in video sequences and plays an important role in digital video compression. But it is computational demanding part of the encoder. Multi-view video is obtained by capturing one three-dimensional scene with many cameras at different positions. ME for Multi-view video requires high computational complexity. To reduce computational complexity and maintain the image quality, a fast motion estimation method is proposed in this paper. The proposed method uses a hierarchical search strategy. This strategy method consists of modified diamond search patten, multi gird diamond search pattern, and raster search pattern. These search patterns place search points symmetrically and evenly that can cover the overall search area not to fall into the local minimum or exploits the characteristics of the distribution of motion vectors to place the search points. Experiment results show that the speedup improvement of the proposed method over TZ search method (JMVC) can be up to 1.2 ~3 times faster while maintaining similar video quality and bit rates.

• Journal of Korea Multimedia Society
• /
• v.15 no.6
• /
• pp.814-826
• /
• 2012

In this paper, we propose scheduling algorithms for transporting variable-bit-rate video stream using playback buffer level of the clients over wireless communication networks. The proposed algorithms attempt to maximize the utilization of limited bandwidth between the central video server and the clients over a mobile network. Since a video server may serve several video request at the same time, it is important to allocate and utilize network bandwidth to serve them fairly and efficiently. In order to improve the quality of service and real-time performance of individual video playback, the video server attempts to allocate temporarily more network bandwidth to serve a video request with the lower buffer level preferentially. The simulation results prove the fair service and load balancing among the mobile concurrent clients with different buffer levels and hence maximizing the number of frames that are transported successfully to the client prior to their playback times.

• Journal of Broadcast Engineering
• /
• v.14 no.6
• /
• pp.703-720
• /
• 2009

This paper proposes a realistic multi-view scalable video coding scheme designed for user's interest in 3D content services and the usage in the future computing environment. Future video coding schemes should support realistic services that make users feel the 3-D presence through stereoscopic or multi-view videos, as well as to accomplish the so-called one-source multi-use services in order to comprehensively support diverse transmission environments and terminals. Unlike the most of video coding methods which only support two-dimensional display, the proposed coding scheme in this paper is the method which can support such realistic services. This paper designs and also implements the proposed coding scheme through integrating Multi-view Video Coding scheme and Scalable Video Coding scheme, then shows its possibility of realization of 3D services by the simulation. The simulation results show the proposed structure remarkably improves the performance of random access with almost the same coding efficiency.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.3
• /
• pp.261-266
• /
• 2020

As pioneers of the early video art, Steina Vasulka(1940-) and Woody Vasulka(1937-2019) had lead not only their own experimental arts, but also entire changes of contemporary avant-garde performance, music, and visual art. Two artists invented and developed electronic machines for video image-processing by collaborating with engineers, and performed creative experiment on transformation of digital image. For them, video art is not just a means of documentation. The Vasulkas' artistic practices were not bounded by conventional canons and rules in art world, and preferably were parts of active aesthetic strategies for coexistence of vision of human and vision of machine. Particularly, their video art recognized the video as the key medium in an era where media technology began to dominate the system of communication, and established artist's authority over manipulation of moving image electronically without depending on video camera. In that regard, we can value on their video art. Therefore, the paper reflects on the Vasulkas' art and life which have not yet been studied, and suggests academic interests in the context of their artistic activities and aesthetic strategies.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.157-164
• /
• 2012

This paper introduces an implementation of SVC player which runs on Android platform and can play SVC video stream on line from SVC video server. SVC(Scalable Video Coding) is a scalable video encoding technique which supports three scalability such as temporal scalability, spatial scalability, and quality scalability. To implement the SVC player on Android, we implemented a SVC decoder using JSVM open source written in C/C++ as a native part on Android and developed Android UI in Java. Also we built an SVC encoding system off line and an SVC streaming server to conduct on-line SVC streaming experiments. Finally, after we installed the SVC player developed in this paper on Motoroi mobile phone, we evaluated and analyzed on-line streaming performance of the SVC player. The result showed that the player worked well and it had no jitter in streaming with the size of QCIF and 10fps from a fully encoded SVC video source.