11月19日
By 1月时特约编辑
专题文章

多路分配器 '18亮点:编解码器和压缩的未来

多路分配器视频工程师年会是视频工程师开的吗. 将于10月17日和18日在旧金山举行, the conference included 31 speakers giving talks in rapid fire fashion ranging in duration from 10 to 30 minutes. 用一个老生常谈但恰当的比喻, the experience is like drinking from a fire hose: almost impossible to comprehend and digest in real time. I attended the first day this year, and watched several talks from the second day via the 视频库在抽搐.tv.

Overall, the videos are an invaluable source of 信息 on a wide variety of topics. 在这个故事中, 我将回顾一些我觉得最有趣的演讲, 这对我来说显然是独一无二的. Before glancing through my list below, I strongly suggest that you review a 演讲者和主题的完整列表. You’ll likely find multiple talks that you’ll want to watch in addition to those that I discuss below, 按呈现顺序列出.

第一天脱毛

The conference started with a bang thanks to a talk entitled "Streaming the 2018 FIFA World Cup Live in UHD with HDR" by Fubo.比利·罗梅罗和托马斯·辛博斯基. The object of the exercise was delivering a four-rung HEVC encoding ladder ranging from 2160p at 16 Mbps to 720p at 3.5mbps，全部使用HDR10 HDR元数据(图1). The entire workflow was cloud-based and involved transcoding the 70 Mbps input feed in the cloud on AWS C5.18个X-Large实例.

图1. Instance details and the encoding ladder used for the FIFA World Cup 4K delivery with HDR10.

在演讲中, the presenters delivered a blueprint for any video engineer seeking to produce a similar experience, 覆盖网络设置的收购, 编码器的选择, 包装与储存, 客户和玩家的考虑, 包括在亚马逊Fire TV/安卓 TV上的经验教训, Roku, Chromecast超, 以及使用ExoPlayer的Apple TV设备, 动静脉, Roku, 沙加, 还有Bitmovin的玩家. 最终, 主持人建议与会者“快速失败”, 快速学习, 关注用户体验.”

The next talk in my wheelhouse was "What to Do After Per-Title Encoding" by Ben Dodson and Nick Chadwick from Mux. 在快节奏的谈话中, Dodson and Chadwick reviewed the history of 每部编码 and many of the foundational theories and challenges. 然后 the pair detailed how Mux built its 自己的每部编码 facility using machine learning, 以及他们如何将标题编码扩展到场景编码, 哪个启用了实时标题编码. This is a dense and technically challenging presentation that anyone designing a per-title or per-scene encoder will find invaluable.

可感知的视频质量是我们工作的核心, and Twitter’s Sebastiaan Van Leuven’s talk "Subjective 视频质量 Assessment for Mobile Devices" tackled this subject head on. 在他十分钟的演讲中, Van Leuven first reviewed two commonly used techniques for measuring video quality, 单刺激和双刺激平均意见评分(MOS). 短暂的, single-stimulus shows a single sample and asks for a rating on a five-point scale, 而双刺激先显示原始视频，再显示编码后的样本, 并要求类似的评级. 虽然部署很简单, 这两种测试方法在精确性和一致性方面得分都很低, both with different testers rating the same video and the same tester rating the same video on a different day.

提高一致性和可靠性, Twitter developed an Adaptive Paired Comparison (APC) that shows two samples and asks the subject which is better, 就像验光师问的那样, 哪个看起来更好?, 左或右?” This test methodology produces more accurate and reproducible results but can also be very time-consuming. What’s novel about Twitter’s approach is the active learning procedure using a particle filtering simulation that streamlines the sample selection. The short presentation provides an overview, which Van Leuven supplemented with a 链接到博客文章.

加速AV1播放与dav1d

The Alliance for Open Media’s (AOM) AV1 codec launched in mid-2018 but hardware-accelerated playback isn’t expected until mid-2020. This makes software decoder efficiency absolutely critical for deployments over the next 24 month. Many initial tests of AV1 decoding using the AOM decoder libaom, including my 自己的结果显示，它的速度缓慢且效率低下. 出于这个原因, AOM sponsored the development of a new open-source AV1 decoder called dav1d by the VideoLAN, VLC, 和FFmpeg社区.

在他们题为“介绍戴维”的演讲中, “一个新的AV1解码器," VideoLAN’s Jean-Baptiste Kempf and Two Oriole’s Ronald Bultje described the goals of the project, 哪一个包含更小的源代码, 更小的二进制可执行文件, 运行时内存占用比libaom小. 在演讲中, Bultje回顾了大卫迄今为止的表现, 并预测，一旦全面实施, 它将产生与H相似的解码性能.264、HEVC和VP9. While this won’t match the decode efficiency of codecs supported in hardware, 它肯定会将AV1的使用范围扩展到libaom无法承受的范围. 根据这个博客帖子, Dav1d目前工作在x86上, x64, v7, ARMv8硬件和运行在Windows上, Linux, macOS, 安卓, 和iOS.

正如RealEye Media的大卫·哈桑在他的演讲中指出的那样, “Multi-CDN Jump Start, 不要把所有的东西放在一个篮子里," using a single CDN to deliver your traffic means a single point of failure, 当流式传输是关键任务时，这是不可接受的风险. 哈桑也提到过, 单个CDN也可能无法为许多用户提供最佳体验, 而且可能不符合成本效益.

这些观点, Hassoun随后指出了使用多个cdn的常见问题, 比如直播的同步起源, 交通路由, 为QoS和QoE接收可操作的实时数据, 以及跨CDN访问安全. 然后, he suggests multiple solutions to these problems and how to build multiple CDN support all the way d自己的 to manifest file creation. 在规定的十分钟内讲了很多内容, this presentation is a must see for anyone considering dipping their toes into multiple CDN delivery (图2).

图2. 添加多个CDN支持一个动态主播放列表.

减少玻璃到玻璃的延迟

Glass-to-glass latency is a consistent concern of many live-event producers. 虽然有几种专有方法可以减少实时延迟, 比如Wowza流媒体云的超低延迟服务, 这可能不适用于大型活动所需的规模.

One solution that’s gaining traction is Chunked CMAF as comprehensively described by Akamai’s Will Law in his presentation entitled "Chunky Monkey, Using Chunked-Encoded Chunked-Transferred CMAF to Bring Low Latency Live to Very Large Scale Audiences.图3说明了这种方法. 最重要的是传递细分市场的传统方式, 什么是等到它完全定型并储存起来. The bottom shows the same media samples packaged in chunks that can be delivered before the complete segment is encoded and saved, 这大大减少了延迟.

图3. 顶部是完成后交付的单个分段. On the bottom are the same samples packaged in chunks delivered chunk by chunk.

尽管这种方法减少了延迟并简化了网络吞吐量, 它还会引发多种问题, like how to estimate bandwidth and how to resolve timing differences between HLS and DASH. Law discussed different solutions to these issues and concluded with a look at standardization efforts for chunked CMAF, as well as commercial vendors and open source tools for implementing this approach.

下一个页面