• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1049-1058
• /
• 2022

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF101) for improvement in prediction accuracy.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.301-304
• /
• 2013

In this paper, to improve the fast boot and soft real-time for the responsiveness in Qplus with relative evaluation rather than an absolute evaluation using validated to perform the services actually used. Qplus is included soft real-time, fast booting, and file system technology. There is a development board that mounted on an embedded Linux and Qplus. AV (Audio / Video) group communications is ported on the development board for soft real-time multimedia streaming services and to evaluate the performance of communication. We performed three major evaluation such as booting time measurement experiment, own performance measurement experiments for soft real-time measurement experiment, and performance measurement experiments for via the network communication performance evaluation. They are a practical Qplus performance evaluation and analysis.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.571-573
• /
• 2014

In today's technology streaming service's QoS technologiey is an issue. On quality video streaming service there are some technical issues exist, such as buffering. This submission is "Adaptive dual-streaming system design" which is for the integrity of the data streaming that is sent to TCP and UDP for faster transmission of data to the stream. This system provides real-time incoming video encoding in bitrate of h.264-based H through a process based on the video footage of several server and client-to-TCP and UDP via Adaptive providing streaming services in a network environment. This is an unspecified number of buffers in a network environment and continued through the minimization of various streaming for playback of videos and multimedia will be utilized in the field.

• Journal of Information Processing Systems
• /
• v.19 no.4
• /
• pp.439-449
• /
• 2023

This paper deals with a resource efficient artificial intelligence (AI) service architecture for multi-channel video streams. As an AI service, we consider the object detection model, which is the most representative for video applications. Since most object detection models are basically designed for a single channel video stream, the utilization of the additional resource for multi-channel video stream processing is inevitable. Therefore, we propose a resource efficient AI service framework, which can be associated with various AI service models. Our framework is designed based on the modular architecture, which consists of adaptive frame control (AFC) Manager, multiplexer (MUX), adaptive channel selector (ACS), and YOLO interface units. In order to run only a single YOLO process without regard to the number of channels, we propose a novel approach efficiently dealing with multi-channel input streams. Through the experiment, it is shown that the framework is capable of performing object detection service with minimum resource utilization even in the circumstance of multi-channel streams. In addition, each service can be guaranteed within a deadline.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.27-34
• /
• 2021

In this paper, we propose a copyright protection scheme for real-time streaming of HEVC(High Efficiency Video Coding) based realistic content. Previous research uses encryption and modular operation for copyright pre-protection and copyright post-protection, which causes delays in ultra high resolution video. The proposed scheme maximizes parallelism by using thread pool based DRM(Digital Rights Management) packaging with only HEVC's CABAC(Context Adaptive Binary Arithmetic Coding) codec and GPU based high-speed bit operation(XOR), thus enabling real-time copyright protection. As a result of comparing this scheme with previous research at three resolutions, PSNR showed an average of 8 times higher performance, and the process speed showed an average of 18 times difference. In addition, as a result of comparing the robustness of the forensic mark, the filter and noise attack, which showed the largest and smallest difference, with a 27-fold difference in recompression attacks, showed an 8-fold difference.

• Applied Microscopy
• /
• v.40 no.1
• /
• pp.47-51
• /
• 2010

The capability of real-time observation in TEM is quite useful to study dynamic phenomena of materials in a certain variable ambience. In performing the experiment, the choice of video recording program is an important factor to obtain high quality of movie streaming. Window Movie Maker (WMM) is generally recommended as a default video recording program if one uses "DV Capture" function in DigitalMicrograph$^{TM}$ (DM) software. However, the image quality does not often satisfy the condition for high-resolution microscopic analysis since the severe information loss in the final result occurs during the conversion process. As a good candidate to overcome this problem, Virtual-Dub is highly recommended since the information loss can be minimized through the streaming process. In this report, we demonstrated how useful VirtualDub works in a high-resolution movie recording. Quantitative comparison of the information quality between the images recorded by each software, WMM and VirtualDub, was carried out based on histogram analysis. As a result, the image recorded by VirtualDub was improved ~13% in brightness and ~122% in contrast compared with the image obtained by WMM at the same imaging condition. Remarkably, the gray gradation (meaning an amount of information) becomes wider up to ~115% than that of the WMM result.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.147-151
• /
• 2014

With the advent of tiny and inexpensive computer systems like Raspberry Pi, it has become possible to construct an experimental environment of MANETs (Mobile Ad hoc Networks) at a relatively low cost. In this paper, we implement a node of MANETs with a Raspberry Pi equipped with two 802.11 wireless LAN cards and dynamic routing software running on it. In an outdoor MANETs environment with moving mobile nodes automatically rebuilding broken routes, we conducted a real-time video streaming experiment over the MANET. The video stream is transmitted from a node with a video camera installed on it to another node through the route established dynamically. Each node is equipped with a display device, with which the current status of the node can be checked out. We also implemented software that automatically collects data stored in each node and analyzes the performance of the MANETs. Through this experiment, we could evaluate the performance of realistic video streaming service in a MANETs environment.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.10C
• /
• pp.1469-1483
• /
• 2004

We propose a packet scheduling algorithms for streaming media. We assume that the receiver periodically reports back the channel throughput. From the original video data, the importance level of a video packet is determined by its relative position within its group of pictures, taking into account the motion-texture discrimination and temporal scalability. Thus, we generate a number of nested substreams. Using feedback information from the receiver and statistical characteristics of the video, we model the streaming system as a queueing system, compute the run-time decoding failure probability of a Same in each substream based on effective bandwidth approach, and determine the optimum substream to be sent at that moment in time. Since the optimum substream is updated periodically, the resulting sending order is different from the original playback order. From experiments with real video data, we show that our proposed scheduling scheme outperforms the conventional sequential sending scheme.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.1
• /
• pp.1-12
• /
• 2003

Recently, many researches into distributed processing on Internet, and multimedia data processing have been performed. Rapid and convenient multimedia services supplied with high quality and high speed are to be needed. In this paper, we design and implement clip-based video retrieval system on the Web enviroment in real-time. Our system consists of the content-based indexing system supporting convenient services for video content providers, and the Web-based retrieval system in order to make it easy and various information retrieval for users in the Web. Three important methods are used in the content-based indexing system, key frame extracting method by dividing video data, clip file creation method by clustering related information, and video database construction method by using clip unit. In Web-based retrieval system, retrieval method ny using a key word, two dimension browsing method of key frame, and real-time display method of the clip are used. In this paper, we design and implement the system that supports real-time display method of the clip are used. In this paper, we design and implement the system that supports real-time retrieval for video clips on Web environment and provides the multimedia service in stability. The proposed methods show a usefulness of video content providing, and provide an easy method for serching intented video content.

• Journal of Communications and Networks
• /
• v.16 no.5
• /
• pp.548-558
• /
• 2014

Load distribution is vital to the performance of multipath transport. The task becomes more challenging in real-time multimedia applications (RTMA), which impose stringent delay requirements. Two key issues to be addressed are: 1) How to minimize end-to-end delay and 2) how to alleviate packet reordering that incurs additional recovery time at the receiver. In this paper, we propose sub-packet based multipath load distribution (SPMLD), a new model that splits traffic at the granularity of sub-packet. Our SPMLD model aims to minimize total packet delay by effectively aggregating multiple parallel paths as a single virtual path. First, we formulate the packet splitting over multiple paths as a constrained optimization problem and derive its solution based on progressive approximation method. Second, in the solution, we analyze queuing delay by introducing D/M/1 model and obtain the expression of dynamic packet splitting ratio for each path. Third, in order to describe SPMLD's scheduling policy, we propose two distributed algorithms respectively implemented in the source and destination nodes. We evaluate the performance of SPMLD through extensive simulations in QualNet using real-time H.264 video streaming. Experimental results demonstrate that: SPMLD outperforms previous flow and packet based load distribution models in terms of video peak signal-to-noise ratio, total packet delay, end-to-end delay, and risk of packet reordering. Besides, SPMLD's extra overhead is tiny compared to the input video streaming.