Advanced User Adaptive Video Demo at Broadband World Forum

Presenter: This is a demonstration of a user-adaptive model, with the [stream end] technology. This technology is based on the understanding that the mobile environment devices such as cell phones or tablets can [be hold] differently. For example, the user might be holding tablet in hand, he might put it in his laps, or he might put it on table, on stand of some sort. And, in all those cases, the distance between device and the viewer might be different.

That distance affects the degree to which a viewer is able to discern information on the screen and in our technology, we take advantage of this different environments and process information in a way that is tailored to each different viewing environment and condition, differently, to maximize the quality and minimize the [lured] on network, needed to deliver that information.

They announced this technology about four months ago, at Mobile World Congress in Barcelona, and today, they are showing four different implementations of this technology on different devices, with different form-factors, different access engines and different operating systems.

For example, among implementations that we are showcasing, today is the streaming client from [Real] Networks. It’s called Helix. And, it’s using user-adaptive functionalities that we integrated in. We also have streaming client from [Idiom], and in this particular case, the [Transit] Android and the Transit extremely high resolution, it’s 14, 14p, it’s two times higher than the standard full HD resolution.

Last, but not least, we are also showing streaming client, which is based on [Dash] Industry for Java Script, dash, access engine. Let me now show how it works in details. To start the stream, we select a video clip and press Load. And, the streaming client starts loading the video and around the screen. Our - back message is saying that we detected that in this particular situation, the device is on the stand and it limits the bandwidth to be used by the client, to four megabits.

Now, if I would pick the device and hold close it to myself, video detects the device is now in-hand, and the bandwidth is increasing accordingly. In this particular example, the bandwidth is increased to 5.7 megabits and it will actually increase to six megabits, which is the highest available bandwidth in our set of representations. Right now, it’s up at six megabits.

If I am, again, putting the device away, it will detect that I am further away from the device, that I am not paying attention to it, and it will reduce the load and will reduce the network consumption and will continue to stream the same clip but at lower rate, which is still sufficient, but uses much less of network.

This technology is intended to benefit a number of players in streaming eco-system. The content publishers can benefit from lower CDM costs as the amount of data that will be consumed by end users will be minimized by this technology. The result of use of this technology allows a lower network load on the access networks, in particular mobile networks, or resources that are shared between all users and by lowering consumption at one point, the bandwidth is reused by other users. So, more users will be able to appreciate the video with high quality, and with less buffering.

And, of course, it will benefit end users as it will allow them to enjoy video with high as possible quality and will minimize their wireless bill.

Advanced User Adaptive Video Demo Interview with Yuriy Reznik

Yuriy Reznik: This is the demonstration of user-adaptive video streaming. The purpose of this technology is to minimize load and network [unintelligible] data traffic. It is of particular importance to wireless network where the capacity is shared within users and several users using [more] [unintelligible] [antennas] [unintelligible]. So, the principles that we are using to reduce this load are based on phenomenon of human vision. Is facts such as the human visual acuity limits as well as [unintelligible] sensitivity curve. In a nutshell, we are using some [sort of] signal device, such as front-facing camera, for a proximity sensors to detect the distance between the human and the screen.

For example, I will ask my colleague Ed to stay close to the screen. And as you can see in this example when he is standing close there is not much that we can [save] because there is a broad range of spatial frequencies that he can see and appreciate while looking at the screen. As Ed moves back we can now appreciate more [savings] by removing the spatial frequencies that are no longer visible by human eye. In this particular example the savings are about 70 percent of the bit rate.

This technology can help [several players] in the streaming [unintelligible] systems; content owners and hosting companies can save money by paying less for distribution of the content over [CDMs]; it could help network operators [unintelligible] reduce [unintelligible] their networks and it will help end users as reduced load will increase changes of them receiving content without interruption.

VIDEOS

Advanced User Adaptive Video Demo at Broadband World Forum

Advanced User Adaptive Video Demo Interview with Yuriy Reznik