• ETRI Journal
• /
• 제46권3호
• /
• pp.538-549
• /
• 2024

A conventional codec aims to increase the compression efficiency for transmission and storage while maintaining video quality. However, as the number of platforms using machine vision rapidly increases, a codec that increases the compression efficiency and maintains the accuracy of machine vision tasks must be devised. Hence, the Moving Picture Experts Group created a standardization process for video coding for machines (VCM) to reduce bitrates while maintaining the accuracy of machine vision tasks. In particular, in-loop filters have been developed for improving the subjective quality and machine vision task accuracy. However, the high computational complexity of in-loop filters limits the development of a high-performance VCM architecture. We analyze the effect of an in-loop filter on the VCM performance and propose a suboptimal VCM method based on the selective activation of in-loop filters. The proposed method reduces the computation time for video coding by approximately 5% when using the enhanced compression model and 2% when employing a Versatile Video Coding test model while maintaining the machine vision accuracy and compression efficiency of the VCM architecture.

• ETRI Journal
• /
• 제30권3호
• /
• pp.441-450
• /
• 2008

The scalable extension of H.264, known as scalable video coding (SVC) has been the main focus of the Joint Video Team's work and was finalized at the end of 2007. Synchronization between media is an important aspect in the design of a scalable video streaming system. This paper proposes an efficient media synchronization mechanism for SVC video transport over IP networks. To support synchronization between video and audio bitstreams transported over IP networks, a real-time transport protocol/RTP control protocol (RTP/RTCP) suite is usually employed. To provide an efficient mechanism for media synchronization between SVC video and audio, we suggest an efficient RTP packetization mode for inter-layer synchronization within SVC video and propose a computationally efficient RTCP packet processing method for inter-media synchronization. By adopting the computationally simple RTCP packet processing, we do not need to process every RTCP sender report packet for inter-media synchronization. We demonstrate the effectiveness of the proposed mechanism by comparing its performance with that of the conventional method.

• 한국통신학회논문지
• /
• 제32권9C호
• /
• pp.834-839
• /
• 2007

The superimposed text in video brings important semantic clues into content analysis. In this paper, we present the new and fast superimposed text localization method in video segments. We detect the superimposed text by using temporal information contained in the video. To detect the superimposed text fast, we have minimized the candidate region of localizing superimposed texts by using the difference between consecutive frames. Experimental results are presented to demonstrate the good performance of the new superimposed text localization algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Journal of Computing Science and Engineering
• /
• 제7권3호
• /
• pp.168-176
• /
• 2013

Long-term evolution (LTE) is emerging as a major candidate for 4G cellular networks to satisfy the increasing demands for mobile broadband services, particularly multimedia delivery. Multiple-input multiple-output (MIMO) technology combined with orthogonal frequency division multiple access and more efficient modulation/coding schemes (MCS) are key physical layer technologies in LTE networks. However, in order to fully utilize the benefits of the advances in physical layer technologies, the MIMO configuration and MCS need to be dynamically adjusted to derive the promised gains of 4G at the application level. This paper provides a performance evaluation of video traffic with variations in the physical layer transmission parameters to suit the varying channel conditions. A quantitative analysis is provided using the perceived video quality as a video quality measure (evaluated using no-reference blocking and blurring metrics), as well as transmission delay. Experiments are performed to measure the performance with changes in modulation and code rates in poor and good channel conditions. We discuss how an adaptive scheme can optimize the performance over a varying channel.

• ETRI Journal
• /
• 제36권4호
• /
• pp.528-536
• /
• 2014

High Efficiency Video Coding (HEVC) is the most recent video coding standard to achieve a higher coding performance than the previous H.264/AVC. In order to accomplish this improved coding performance, HEVC adopted several advanced coding tools; however, these cause heavy computational complexity. Similar to previous video coding standards, motion estimation (ME) of HEVC requires the most computational complexity; this is because ME is conducted for three inter prediction modes - namely, uniprediction in list 0, uniprediction in list 1, and biprediction. In this paper, we propose an efficient inter prediction mode (EIPM) decision method to reduce the complexity of ME. The proposed EIPM method computes the priority of all inter prediction modes and performs ME only on a selected inter prediction mode. Experimental results show that the proposed method reduces computational complexity arising from ME by up to 51.76% and achieves near similar coding performance compared to HEVC test model version 10.1.

• 대한산업공학회지
• /
• 제25권2호
• /
• pp.162-172
• /
• 1999

For VBR video traffic, Motion-Picture Experts Group(MPEG) coding algorithm was adopted as the standard coding algorithm by International Telecommunication Union(ITU). In this paper, we propose a traffic model of an MPEG coded video traffic in frame level and cell level, and develop an approximate model for evaluating performance of a ATM multiplexer with homogeneous MPEG video sources by considering burst-level variation of aggregated traffics. For homogeneous MPEG video traffics which are frame-synchronized, the performance of the ATM multiplexer is influenced by source correlation at the multiplexing time. When sources are highly correlated, we decompose the aggregated cell streams by the frame-type and model multiplexing process during a frame time as n*D/D/1/K queueing model and suggest an approximate method for obtaining CLP of the ATM multiplexer. In the case that sources are highly correlated, the solution has the meaning of the upper bounds of performance of the ATM multiplexer. For the verification of our model, we compare the solution of our model with simulation resets. As the number of sources increases. The CLP obtained from our model approaches to simulation results, and gives upper bounds of simulation results.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제20권11호
• /
• pp.586-591
• /
• 2014

최근 언제 어디서든 동영상 컨텐츠에 접근할 수 있게 됨으로써 배포된 영상은 쉽게 복사되고, 변형되고 재배포 되어 저작권 보호에 취약하다는 문제점을 내포하고 있다. 따라서 비디오 복사본의 유사도를 검출하고 측정하는 방법들이 요구되어진다. 본 논문에서는 복사본 검출 기술 중 MPEG에서 표준화 한 MPEG-7 Video Signature를 이용하여 다양한 변화를 갖는 동영상에서의 다양한 분별력 분석을 하였다. MPEG-7 Video Signature는 블록기반의 추상화 방식이므로 동영상의 영역 변화에 대해서 취약할 것이라고 가정하고 실험하였다. 분석한 결과 실제로 영역변화에 대해서 일반적으로 일어날 수 있는 강도에서도 매우 취약함을 볼 수 있었다.

• 인터넷정보학회논문지
• /
• 제15권3호
• /
• pp.45-52
• /
• 2014

유비쿼터스 시티(유시티)에서는 수많은 비디오 카메라들이 설치된다. 이렇게 설치된 많은 카메라로부터 대용량의 비디오 데이터가 실시간으로 끊임없이 발생하고 유시티의 관리 시스템으로 전달된다. 유시티의 다양한 서비스들을 뒷받침하기 위해서는 이러한 비디오 데이터를 저장하고, 이렇게 저장된 대용량의 비디오 데이터를 분석할 수 있는 방법과 관리 시스템이 요구된다. 그래서, 이 논문에서는 클라우드 컴퓨팅을 기반으로 한 유시티 비디오 관리 시스템을 제안한다. 또한, 근래 주목받고 있는 데이터 병렬처리 프레임워크인 Hadoop MapReduce를 이용하여 이러한 빅데이터 비디오를 분석하는 방법을 제안하고, 이에 따른 우리의 성능 평가를 소개한다.

• 방송공학회논문지
• /
• 제18권6호
• /
• pp.872-883
• /
• 2013

본 논문에서는 패킷 오류가 발생하는 IP망을 통해 SVC 비디오 전송 서비스를 제공하기 위한 Exclusive-OR 기반의 FEC (forward error correction) 오류제어 시스템을 설계하고 성능을 분석한다. 설계된 시스템에서는 계산적으로 복잡도가 낮은 표준 Exclusive-OR 연산에 기반한 FEC 방법을 활용하고, SVC 비디오의 계층적 구조에 적합하도록 FEC 기법을 적용 한다. 설계된 Exclusive-OR 기반의 오류 제어 시스템의 성능을 검증하기 위하여 NIST-NET 기반의 전송 시뮬레이터를 활용한다. NIST-NET 기반의 시뮬레이터를 통한 SVC 비디오 패킷 전송 실험에 의해 설계된 Exclusive-OR 기반의 FEC 시스템의 오류 강인 전송 성능을 확인한다.