TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Stanford Researchers Launch Free TV Service to Improve Live Streaming Using AI

114 点作者 fomopop超过 6 年前

12 条评论

keithwinstein超过 6 年前
Hi folks -- study PI here. Happy to answer any questions.<p>We&#x27;re working on a paper for submission to a research conference soon, so every viewer gives us helpful data for the system to learn in-situ and improve its performance over different kinds of Internet connections and congestion situations. (We are randomizing connections to get either our in-development algorithm or a bunch of competing ABR and congestion-control schemes so we can properly evaluate, so if you get a stall or bad quality...... we&#x27;ll blame that.)<p>All the source code is at <a href="https:&#x2F;&#x2F;github.com&#x2F;StanfordSNR&#x2F;puffer" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;StanfordSNR&#x2F;puffer</a> and is or will be released with an open-source license. The study is led by my doctoral student Francis Yan, along with Sadjad Fouladi, Hudson Ayers, Chenzhi Zhu, and my colleague Philip Levis. The goal is to train video-streaming algorithms (bitrate selection and congestion control) online, continually, in situ, to minimize stalls and deliver the best picture quality across a diverse range of real-world Internet connections and users. This is from some of the same people who brought you Salsify (functional video compression for lower-latency real-time video: <a href="https:&#x2F;&#x2F;snr.stanford.edu&#x2F;salsify" rel="nofollow">https:&#x2F;&#x2F;snr.stanford.edu&#x2F;salsify</a>), Lepton (distributed&#x2F;parallel JPEG compression: <a href="https:&#x2F;&#x2F;github.com&#x2F;dropbox&#x2F;lepton" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;dropbox&#x2F;lepton</a>), Mosh (functional terminal emulation for mobility: <a href="https:&#x2F;&#x2F;mosh.org" rel="nofollow">https:&#x2F;&#x2F;mosh.org</a>), Remy (machine learning for congestion control; <a href="http:&#x2F;&#x2F;mit.edu&#x2F;remy" rel="nofollow">http:&#x2F;&#x2F;mit.edu&#x2F;remy</a>), and the Pantheon of Congestion Control (<a href="https:&#x2F;&#x2F;pantheon.stanford.edu" rel="nofollow">https:&#x2F;&#x2F;pantheon.stanford.edu</a>).<p>There are some unusual details of the Puffer system to make experimentation easier (and, we hope, performance better): the ABR algorithm is server-side, and video is streamed continuously over a WebSocket and the client gives asynchronous feedback, instead of using DASH HTTP request&#x2F;replies. The tcp_info struct (including the delivery rate estimate) from the congestion-control layer is plumbed through to the ABR, and to the neural network that tries to predict &quot;how long will it take to send a video chunk of length y,&quot; so it&#x27;s weakly cross-layer. The &quot;transmission-time predictor&quot; is probabilistic instead of simply producing a point estimate. We encode with libx264 and then measure the SSIM of every encoded chunk to calculate a reward for each possible stream (instead of making the assumption that more bitrate is better -- the correlation is not so good when you consider different times in the same video). And we encode 10 variants of each channel, so more than typical. And everything is done at 60 fps. Of course some of these details mean these algorithms are not going to be deployed any time soon on a production CDN or streaming service, but the hope is to demonstrate the value of some of the ideas in an academic setting for future commercial deployment. The use of WebSockets, MSE, and Opus is why this doesn&#x27;t work on Safari or on iOS.<p>FAQ here: <a href="https:&#x2F;&#x2F;puffer.stanford.edu&#x2F;faq&#x2F;" rel="nofollow">https:&#x2F;&#x2F;puffer.stanford.edu&#x2F;faq&#x2F;</a><p>Also I&#x27;m happy we were able to make some of the Grafana monitoring public: <a href="https:&#x2F;&#x2F;puffer.stanford.edu&#x2F;monitoring" rel="nofollow">https:&#x2F;&#x2F;puffer.stanford.edu&#x2F;monitoring</a>
评论 #18938655 未加载
评论 #18937943 未加载
评论 #18937380 未加载
abhiminator超过 6 年前
Streaming it in another browser window as a part of my Mechanical Turk HIT assignment as I type, and I have to say -- I&#x27;m blown away by the incredible picture quality and buffer-free performance.<p>Puffer team has done a great job.<p>As a sidenote, I just realized how annoying cable TV ads are. The content to advertising ratio on cable is insanely bad. This &#x27;Puffer&#x27; experience just validated my decision to stay away from cable TV.
评论 #18937465 未加载
daveevad超过 6 年前
Anyone with legal knowledge about how services like this and <a href="https:&#x2F;&#x2F;www.locast.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.locast.org&#x2F;</a> are able to operate?<p>It doesn&#x27;t seem significantly different to me than the aereo tv situation.<p>Very cool project!
评论 #18937074 未加载
评论 #18939257 未加载
phillipseamore超过 6 年前
This is very impressive. FYI: Delaying the buffer spinner, not showing it unless buffering has been (let&#x27;s say) &gt;300ms for instance, makes the switching feel a lot faster. If you could also get rid of the black between switching out the media source buffer, they would feel instant.
milankragujevic超过 6 年前
I&#x27;m going to be experimenting with the source code, I really need some good basis for live streaming TV over the internet (OTT). Unrelated, but important for me, TV stations in my country (Serbia), forbid rebroadcasting but they are available OTA, with a DVB-T2 receiver and an aerial&#x2F;antenna. However, some cable providers cannot show the programs, and if they do show them, they have negotiated special licenses for that. How do I make a case that by being OTA Free, and not DRM&#x27;d, I, by not changing the contents of the broadcast, should be allowed to rebroadcast to people for no financial gain to myself or the TV stations (other than the ads the TV stations put inside of the broadcast content) (all to serve as an experiment)?
sand500超过 6 年前
This is a pretty cool idea<p>Note, you still have to deal with TV ads and there is no scrubbing. I wonder if you could make your own client to this and get DVR functionality? Actually since they are just grabbing over-the-air signals, I wonder if there is a SDR project for this.
bane超过 6 年前
This is actually faster at channel switching than my home fiber TV service.
voltagex_超过 6 年前
It&#x27;s quite nice, but it&#x27;s definitely going to get a DRM layer and geoblocking applied to it at some stage.
评论 #18937007 未加载
homero超过 6 年前
The quality and speed is incredible
voltagex_超过 6 年前
Edit: bitrate and resolution varies - I&#x27;ve seen bitrates as high as 20 megabit&#x2F;s.
masonic超过 6 年前
Is closed captioning supported?
compsciphd超过 6 年前
wondering what the odds are they get a takedown from the NFL during the super bowl or preemptively won&#x27;t show CBS during that time period?