基于A3C的无线异构网络自适应视频流传输控制方法

比特率自适应（ABR）算法已经成为视频传输中研究的热点之一。然而,由于5G无线异构网络具有信道带宽波动大、不同网络间差异明显等特点,多终端协同的自适应视频流传输面临着巨大挑战。提出了一种基于深度强化学习的自适应视频流传输控制方法。首先,建立了视频流动态规划模型,对传输码率以及分流策略进行联合优化。由于该优化问题的求解依赖于精确的信道估计,这在信道状态动态变化的网络中很难实现。因此,将动态规划问题改进为强化学习任务,并采用A3C算法,动态决策视频码率和分流策略。最后,根据实测的网络数据进行仿真,与传统的优化方法相比,本文所提的方法较好地提高了用户QoE。     

基于熵编码的立体视频加密与信息隐藏算法

针对立体视频的安全性,该文提出一种基于熵编码的立体视频加密与信息隐藏算法。首先,结合立体视频编码结构,分析误差漂移的物理机制,并根据立体视觉掩蔽效应,确定左右视点的加密帧和隐秘信息待嵌入帧。其次,在基于上下文自适应二进制算术编码(CABAC)的熵编码中,通过等长码字替换技术,实现立体视频的加密和信息隐藏。实验结果表明,视频码流经加密与信息隐藏之后格式兼容、比特率不变,视频感知质量无明显下降,在计算复杂度和码率增加率上有显著优势。     

多视点立体视频传输系统与错误隐藏算法设计

针对多视点立体视频数据量大、网络带宽受限等问题,提出了一种分组传输多路复用的多视点立体视频实时传输系统。系统将多视点数据通过H.264/AVC压缩编码后,在两个IP链路中进行分组复用,实现了立体视频在带宽受限网络中的实时传输;同时,为了解决IP网络传输中的丢包问题,提出一种联合时域视点间预测的错误隐藏算法。最后通过实验表明,本文错误隐藏算法可以提高多视点立体视频的解码质量。     

多视点纹理加深度编码的联合码率控制方法

码率控制技术是多视点视频编码和传输中一个关键的问题。为了提高三维(3D)视频的整体显示质量,包括虚拟视点质量和编码视点质量,提出一种多视点纹理加深度编码的联合码率控制方法。该算法研究了纹理和深度的关系,采用基于模型方法确定最优的纹理和深度之间的码率比例。根据各个视点编码结果的统计规律,不同的视频序列采用不同的视点间比特分配比例。实验结果表明,与目前流行的多视点码率控制算法相比,该算法在计算复杂度基本保持不变的情况下,平均码率控制误差在0.6%以内,客观质量PSNR最高可提高0.65 dB。     

基于RTP的MPEG-4视频传输策略研究

在简述实时传输协议（RTP）原理及MPEG-4视频编码标准的基础上，提出了MPEG-4视频传输模型，并详细介绍了基于RTP的码率自适应调整算法，通过丢包率和传输延时来分析当前网络状况。采取适当的码率调整策略，使整个传输过程既能充分利用带宽又不引起网络拥塞。实验表明本方案能够在变化的网络状况下取得理想的图像传输质量。     

混合网络下可分级视频的跨层码流提取优化算法

针对混合网络中传输带宽受限以及误码率较高的特点,文章在可分级视频编码的基本码流提取方法基础上,提出一种新的基于应用层和传输层信息的码流提取跨层优化算法。文章还进一步结合信道获取传输带宽等信息以及在应用层得到提取网络抽象层单元的码率以及丢弃单元造成的效应,来保证视频传输的服务质量。     

基于跨层码率适配和差错控制的3D视频无线传输方法

针对纹理视频加深度序列的3D视频无线传输,本文提出一种基于跨层优化的码率适配和差错控制方法.通过最小化端到端3D视频失真,均衡调整和配置应用层3D视频编码的码率和帧内编码更新比例,以及物理层的调制和编码模式,达到信源码率适配信道带宽以及应用层差错控制和物理层信道保护强度相互平衡的目的,进而提高接收端的3D虚拟视点视频质量.实验结果表明,提出的方法能有效的提高3D视频无线传输的性能.     

基于3D-SPIHT的立体视频图像压缩编码

该文提出一种新的立体视频编码方案:在辅助序列中进行视差补偿预测和三维等级数集合分区(3D-SPIHT)编码,3D-SPIHT算法建立在真三维小波分解基础上,通过定义一种新的时空方向树结构,实现了静止图像SPIHT算法的三维扩展,实验结果表明该方案的编码性能略高于传统方案,具有较低的计算复杂度,所产生的嵌入式辅助序列码流,可根据通道带宽自适应调整输出码率,最大限度地提高辅助序列的质量。     

面向立体视频传输的右视点图像错误隐藏

利用立体视频序列视点间相关性及单视点内相关性,提出了一种面向立体视频传输的错误隐藏算法.从基于H.264/AVC立体视频编码结构出发,推断出受损块的参考模式;然后基于出错块的内容特征,根据块视差活力度(TDA)或块运动活力度(TMA)的大小,内容自适应地选择恰当的视点间及时域错误隐藏方法对受损块进行错误掩盖.实验结果表...     

双目立体视频最小可辨失真模型及其在质量评价中的应用

该文针对目前最小可辨失真(JND)模型只能应用于评价单视点视频的问题,探索一种以双眼亮度关系为基本出发点的双目立体视频的JND模型。首先根据立体视频特性引入双眼亮度关系模型,然后考虑了背景亮度掩盖、纹理掩盖、帧间掩盖、空间时间对比灵敏度、眼睛运动等因素,提出了双目JND模型,最后将该双目JND模型应用到立体视频质量评价方法中。实验结果表明该文提出的质量评价方法符合人类的视觉感知特性,更接近于主观测试结果。     

Server-Aided Adaptive Live Video Streaming Over P2P Networks

Three factors, including churn of peers, high transmission delay, and high bandwidth heterogeneity, jointly bring forward great challenges to video streaming over P2P networks. In this paper, the multi-tree approach is leveraged to construct an overlay with resilience to churn and low transmission delay. For such a multi-tree structured overlay, a server-aided adaptive video streaming scheme is proposed to cope with the bandwidth heterogeneity. During streaming process, video data are collaboratively forwarded to the same receiver by multiple peers based on side information and network condition, as well as the distributed bitstream is dynamically switched among multiple available versions in a rate-distortion optimized way by the streaming server. Simulation results show that the proposed scheme achieves great gain in overall perceived quality over simple heuristic schemes.     

A Novel End-to-End Architecture for H.264 Video Streaming over the Internet

In this paper we propose a novel end-to-end architecture for H.264 unicast video streaming over the Internet. The proposed video streaming architecture is based primarily on a new transport layer protocol, the stream control transmission protocol (SCTP). We show that the network-friendly specification of H.264, and the novel technical characteristics of SCTP, when coupled together are able to provide a highly adaptive and flexible system for unicast video streaming. More specifically, we develop algorithms that handle at the transport layer the following functions concerning video packet transmission: retransmission policy, packet prioritization, implicit receiver feedback. We combine the above algorithms with an R-D optimization strategy at the encoder, that provides more options when adapting to bandwidth variations. This combined optimization strategy leads to more options concerning the streaming parameters. Finally, with simulation results we prove that our system is capable of maintaining good perceptual quality and TCP-friendliness under various loss conditions. An early version of this paper appeared in CCNC 2004.     

Q2ABR: QoE‐aware adaptive video bit rate solution

In this paper, we propose a new adaptive bit rate (ABR) streaming method. This method is based on estimating and monitoring users' video streaming experience, their quality of experience (QoE). This ensures a good user QoE and optimises bandwidth utilisation by monitoring video buffer fill rate to ensure minimal data traffic. First, we achieve a QoE evaluation model based on network bandwidth, video segment representation, and dropped video frame rate parameters. Second, following our QoE evaluation model, we formulate an ABR method using the reinforcement learning (RL) paradigm to select video representations and using a breakpoint detection mechanism to monitor end‐user QoE variation. The proposed ABR method is called “QoE‐aware adaptive bit rate (Q2ABR)” and is composed of three individual modules, one for QoE estimation using machine learning methods, one for QoE variation monitoring using the breakpoint detection mechanism, and one for video representation selection using reinforcement learning. The design objective of Q2ABR is to ensure the overall QoE of these users while maintaining a minimum variation in the standard deviation of the users' QoE values. Third, the performance of the Q2ABR method is evaluated and compared with several existing ABR approaches in the literature using real traces that we collect on different transport scenarios (such as bus and train, among others). Since this method considers the user's perception of video quality as a regulator for optimising the overall video distribution network, good results are ensured in terms of the user's experience and buffer fill rate.     

TRASS: A transmission rate‐adapted streaming server in a wireless environment

The video streaming quality in a wireless communication network environment is largely affected by various network characteristics, such as a limited channel bandwidth and a variant transmission rate. The playback quality of User Equipments (UEs) may not be smooth when the service is delivered via a wireless environment. From the viewpoints of most video receivers, a smooth playback with a lower video quality may be more significant than a lagged or distorted playback with a higher video quality as the transmission rate degrades. Based on the above, we sketch an adaptation agent—Transmission‐Rate Adapted Streaming Server (TRASS), which is located between the original video server and UEs, to adaptively transform the streaming video based on the real transmission rate. In our proposed scheme, UEs would feedback their network access statuses to TRASS and then TRASS would deliver adaptive quality of video streams to UEs according to their feedbacks. The theoretical analysis and simulations using different video tracks encoded in MPEG‐4 and H.264/AVC formats show that TRASS can help wireless streaming users to get a smooth playback quality with a lower packet failure rate. With a low probability of receiving a worse quality of video, users' Quality of Experience can subsequently be raised. Copyright © 2010 John Wiley & Sons, Ltd.     

基于三维感知的立体虚拟视点图像质量评价方法

为了评价立体虚拟视点图像的质量,提出了一种基 于三维感知的客观评价方法。综合考虑了立体虚拟视点图像两大最主要失真类型:单视点绘 制失真和立体视点不匹配失真。针对单视点绘制失真,先提取 当前视点失真图与无失真图的差异性区域,再针对该差异性区域计算平均结构相似度(MSSIM ),最后将左 右视点平均池化作为单目纹理特征值;针对立体视点不匹配失真,先对左右视点失真图分别 进行视差映射, 再提取映射图与该视点失真图的差异区域作为双目不匹配区域,然后针对不匹配区域计算MS SIM 值,最 后将左右视点平均池化作为双目竞争特征值;最终将两个特征值幂次融合,作为最终的立体 虚拟视点图像 质量评价客观指标。实验结果表明本方法有效匹配主观打分的DMOS值,皮尔森线性相关系数 和斯皮尔曼 秩相关系数分别为0.911和0.900,正确反映了 立体虚拟视点图像质量。     

一个基于速率控制的Internet视频流服务方案

由于视频流服务对于网络服务质量有着较高的要求,而现有的Internet所提供的是尽力而为的服务,无法保证数据的实时传输。该文设计了一个用于Internet上视频流的端到端传输方案.整个方案设计的目的是在网络本身缺乏服务质量保证的条件下尽可能达到最好的视频传输质量。根据可用带宽估计和网络信息反馈,系统对发送速率进行调整,并提供两种视频流服务:存储视频和实时视频。仿真结果表明方案的性能良好,能满足Internet视频流的需求。     

Joint source and sending rate modeling in adaptive video streaming

This work addresses the modeling of traffic generated by a video source operating in the context of adaptive streaming services. Traffic modeling is a key in several network design issues, such as dimensioning of core and access network resources, developing pricing procedures, carrying out cost-revenue studies. The actual traffic generated during a video streaming session depends on both the video source and the bandwidth variations imposed by lower communication layers. We propose a new traffic model that jointly encompasses these two effects. Specifically, we consider the modeling of the sequence of frame sizes generated by a video streaming source that dynamically adapts its rate to the available communication channel bandwidth using bitstream switching techniques. In order to represent the source rate adaptation to the random network bandwidth variations on the communication channel, we resort to a framework based on Hidden Markov Processes (HMPs). Our HMP model represents the first joint source and sending rate model in adaptive streaming literature. Thanks to effective modeling assumptions on the frame size probability density function (pdf), the HMP parameters can be estimated by means of the Expectation Maximization algorithm. The traffic model is validated by numerical simulations of a mobile adaptive video streaming scenario. We study the model's ability to predict several traffic statistics, including the traffic load of a video streaming source in different network points. Besides, we evaluate the model accuracy in characterizing aggregate video traffic resulting from multiplexing various video sources. In all experiments, we show that the proposed model is able to accurately capture the traffic characteristics.     

Smoothing transcoded MPEG-1 video streams for Internet transmission

An efficient smoothing scheme for the real-time transmission of MPEG-1 transcoded video over 'best-effort' IP networks is presented. The scheme uses intelligent partitioning and multiplexing of the packetised bit stream. Bit-rate smoothing is achieved by partitioning packets according to their picture type (I, P or B). Subsequently, the partitioned packets are multiplexed in such a way that each packet from an anchor (I and P) picture is followed by two packets from B-pictures. The proposed scheme smooths the bit rate of the encoded video, making it more suitable for adaptive video streaming applications. In such applications, the transmission bit rate is varied so as to adapt to the available network bandwidth. The scheme reorganises the transmission order of the packets, spreading the less important packets from the B-pictures into the more important packets from the anchor pictures. This indicates a reduction in the likelihood of losing more important anchor packets as the bandwidth bottleneck increases, implying an improvement in the quality of transmitted video. A variety of simulations results are presented to demonstrate these points.     

基于无线视频服务器自适应带宽算法

针对无线网络上行带宽有限的情况,提出了无线视频传输带宽的自适应算法。采用双卡发送采集的视频流数据,这样大大增加了无线视频传输带宽。解决了公共无线网络带宽资源有限的问题。使得无线视频传输码率能够达到500～900 kbit/s。在接收端采取双缓冲区的设计,在客户端能够得到清晰、流畅的视频图像。从而解决了无线视频传输和带宽不足的问题。