SMW 17: Beamr's Dror Gill Talks Integrated Video Encoding and Optimization
Tim Siglin: Welcome back to Streaming Media West 2017. I'm Tim Siglin, Contributing Editor with Streaming Media, as well the Media Strategy Principal at ReelSolver, Incorporated. Today, I have with me Dror Gill, who's the CTO of Beamr. So, Dror, first of all, before we go into some of the topics, tell me a little bit about Beamr.
Dror Gill: Beamr is the leading provider of video encoding and optimization solutions, and mainly our customers are MSOs, telcos, and OTT providers. We provide the basic HEVC and H.264 encoding solutions, which are very high density and high quality. We also provide what we call video optimization solutions, which are pieces of software that take existing encoded video files and make the bit rate lower without hurting their quality.
Tim Siglin: As a matter of fact, I worked at a company called EuclidIQ for a brief period of time, as their Strategy Officer, so I was very familiar with Beamr from that standpoint. You all acquired Vanguard Video last year?
Dror Gill: Yes. Last year we acquired Vanguard Video. We wanted to have tighter integration between the codec and the optimization software. So we went out, and we looked at 10 different companies, and we found out that Vanguard was the best video encoder out there. We acquired the company in March 2016, and since then we've been working very closely with them on integrating the encoding technology with the optimization technology.
Tim Siglin: So, that ability to go really deep down into the encoder and add the optimization in, at that level. I heard some really good news out of IBC about you all, that people were saying that integration has actually yielded significant benefits, both on the AVC side and the HEVC side.
Dror Gill: Right. We've seen bit rate reductions of around 20 to 50%, typically, on content. Back at NAB, we announced this product, the Beamr 5x HEVC encoder with optimization built in, and it's been doing very well.
Tim Siglin: And your basis, when you say 20 to 50%, that's against something like x265, or ... what's your baseline on that?
Dror Gill: Yeah, this is when you compare it to any other encoder, working in a regular VBR mode. In VBR mode, you set the target bit rate, that's the average bit rate you will get. In our encoder, there's an additional mode called CABR, content adapted bit rate. When you run that, you set the target bit rate, but the actual average will be lower, and this is based on content. We do content adaptive encoding at the frame level. For each frame, we analyze what type of content and how many bits we should allocate, and that's how we can get a very low bit rate on each one of the titles.
Tim Siglin: Right. In fact, I just had David Sayed from Brightcove, who was talking about context-aware encoding. Yuriy Reznik from InterDigital has joined them, and it seems like there's this ... Between, you know, when I was at Euclid, with Beamr, with what Yuriy's doing, there's really a move to do video optimization, and I've wanted this for a decade or more, and I'm glad to see that we're finally there.
Dror Gill: I'm also very happy that it's finally happened. We've been the first. We’ve done video optimization since 2009. We started with JPEGs. And in 2012, or '11, we started with H.264 optimization. Now, the whole industry finally understood the message. It doesn't make sense to encode all of your videos at the same bit rate, because each video has different requirements. And now you see that Netflix are doing it, and everybody's into this per-title, per-scene optimization.
Tim Siglin: I think we're sort of half way through the continuum, because we've moved to per-title. I really want to see us at per-scene, or per-shot, because even within the same title, we have so many different types of content. So if you'll champion that for the future not just of codecs, but of encoding, that'd be really helpful.
Dror Gill: Yes. And we're one step ahead of that, because we're already doing per-frame. Inside the scene, we're adapting to each frame of the video, and that's been our approach from the beginning. This is really how you can really get the optimal bit rate for each piece of content. In some cases, we saw that customers sent us a reel of different types of content concatenated together. If you have a naïve method that just looks at the beginning and tries to estimate what it is, it will fail. But if you look at each frame, and try to find the right encoding parameters and bit rate, then you can really adapt to the changing scene.
Tim Siglin: And obviously you want some level of inter-frame benefit as well ...
Dror Gill: Exactly.
Tim Siglin: ... but if you can do intra-frame, as you say, that's your starting basis.
Dror Gill: Right. You have to look at, at least at pairs of frames because preserving the temporal flow is very important.
Tim Siglin: Right. Yeah, spatial and temporal, you have to deal with both of them. So the two panels you’re on today--you said you moderated one on HEVC implementations. And then the future of video codecs. I think somebody described it to me as the HEVC-AV1 Smackdown. What's your take on HEVC, AV1? Obviously, you own a company that has HEVC compression.
Dror Gill: Right. I was in the audience of a third panel that also talked about HEVC vs. AV1. In that panel it was more of a battle between them. In the panel that I participated in, it was more about encoding techniques and streaming and things like that. We've been big believers in HEVC from the beginning. And actually, the company we acquired, Vanguard, has been developing HEVC for the past five years. And one of the largest--I think the largest--OTT provider, it's well-known, is using our codec for all of their 4K HDR series.
We continue to push forward in HEVC because we believe that this is the way to go in terms of standard. The bit rate efficiencies are there, the quality is there. The market has not advanced as quickly as we thought, and it's been hampered mainly by issues of royalty uncertainties.
Tim Siglin: It's royalty uncertainties, but I also get a sense that, especially with context-aware encoding, AVC is good enough, with the exception of UHD or 4K. It's good enough, and if you can get some of that extra benefit, the workflow stays the same.
Dror Gill: When we were pushing content-adaptive encoding, we were saying, you can reduce 20 to 50% of your H.264 bit rate, so you don't need to move to HEVC, and now that you have HEVC, we know that with HEVC you can save like 30, 40%, but with content-aware HEVC, which we have in our Beamr 5x product, you can save an additional 30 to 40% on top of that. That’s because--and this is something that came up in the panel that I moderated--with HEVC, there really is not any perceptual tool of defining the standard. So this is open to implementation. The efficiency you get in HEVC is because of better mathematical models, compression block sizes and stuff like that. But the basic perceptual principle of optimizing video and adapting it to the content, works for HEVC as well as it does for H.264.
Tim Siglin: Because, as you say, there was nothing added into the standard that was different from what was in AVC.
Dror Gill: Exactly. That's why our content-adaptive works on HEVC, as well. Now we're pushing HEVC because we want to say, okay, you can get double benefit, HEVC and content adaptive.
Tim Siglin: Well, that's fascinating to hear, because I'd wondered, as we went through the period of time with AVC reduction, whether we would get similar scenarios in HEVC, and to hear you say you're getting both the effect of HEVC, as well as the optimization on top of that, that's a very compelling argument.
Dror Gill: I would even say that with HEVC today, we get better savings. And the reason is that H.264 implementations, such as the open source x264, have been optimized and developed over the years, and reached a very high level. With HEVC, the implementations are only starting--
Tim Siglin: Right, x265 is an example.
Dror Gill: Right. And it's behind, so that's why you can still save more ... Today, there's more redundancy left, because the codec itself, the implementation, has not been optimized. But, even on the optimized H.264, we can save a lot, and I believe this will continue with HEVC. So the debate of which codec to use, I think maybe last year, at this show, it was still open. But this year, in April, Samsung announced their support. In June, Apple announced their support. Now iOS 11 is out. By the end of this year, we'll have hundreds of millions of devices supporting HEVC in the field, all of the 4K TVs of course, the Edge browser in Microsoft. If you have HEVC support in the Skylake chip on your PC, supports HEVC. I believe that next year, 2018, will be the year that we'll start to see massive deployment of HEVC in the field, to televisions, to mobile devices, in all of the OTT streaming services. We're already starting to see demand from our customers, more demand for the HEVC codec.
Tim Siglin: And just a final thought on that: This is why, from an AV1 standpoint, it's so compelling for them to get what they have, out into the market, before HEVC becomes de facto. If you're thinking from their standpoint. They have a window of time in which they can go out and argue against it, but the optimization piece that you're putting in there makes the argument even more difficult.
Dror Gill: Right. Also, if we talked about maturity of codecs, HEVC, the software implementation, we've been developing for the past five years. AV1 is going to freeze, as we heard at the show, in the next one or two weeks, but it is a reference implementation, not an optimized one. So, sure, you can spin up to 100 servers on the cloud and do one stream, that's nice for a demo, but that's not realistic to do tomorrow, or in 2018, for live streaming.
Tim Siglin: All right. Well, I'm going to stay on the fence between the two, but it's very good to hear from you, the optimization piece, having additional benefit in HEVC beyond what we saw in AVC. Dror, I really appreciate your time.
Dror Gill: Thank you very much.
Tim Siglin: Absolutely fascinating.
Beamr's Dror Gill makes the case for widespread HEVC implementation now.
Streaming Media's Tim Siglin interviews VideoRx's Robert Reinhardt at Streaming Media West 2017.
Streaming Media's Tim Siglin interviews Brightcove VP of Product Marketing at Streaming Media West 2017.
Save on bandwidth and keep costs down: Per-title video encoding solutions let publishers break free from fixed encoding ladders. Explore the benefits of four methods.