DVB takes first step towards possible Virtual Reality delivery spec

By Barry Flynn, Contributing Editor

David Wood, Chair of the DVB’s new Study Mission Group on Virtual Reality (VR), thinks it’s “quite possible” that once the Group has reported back on the results of its research in June, DVB members “will want to set up a structure to prepare specific requirements – and then a specification for – a VR delivery format.”

Speaking after DVB revealed the creation of the initiative in the pages of its house magazine, DVB Scene, Wood said it had been clear by IBC last year that DVB needed to study VR to decide whether it should be doing something about it.

“It’s impossible to ignore the VR tsunami in the industry,” Wood argues. “But also there are those who claim we should be cautious because of what happened with stereoscopic [i.e. 3D] TV.”

That’s why the new group was asked last autumn “to take a balanced and sober look at the prospects for commercial success with VR, as far as it might affect DVB members,” says Wood.

Wood’s view is that, currently, “it seems more likely that internet streaming will be needed to deliver VR services” than a broadcast stream, but that this could still offer shared experiences, “such as in the coverage of sports or other stadium events. In fact, they may be the lifeblood of VR,” he suggests. “VR allows you to ‘be in the best seat in the stadium` and this is exactly where sports fans want to be. The issues to resolve are how to bring to the display the viewer’s ‘region of interest’ at a manageable bit-rate.”

The ‘region of interest’ or ROI is the portion of a 360-degree view that a VR user chooses to look at at any given moment, which needs ideally to be available instantly to give a viewing experience akin to natural viewing.

Wood suggests the VR content could be created using a 360-degree ‘camera’ – made up of a series of cameras pointing at different angles whose HDTV or UHDTV images would then need to be stitched together.   

“There is still discussion about whether something less than 360 degrees is needed, or indeed whether it to should be a ‘donut`, or a `sphere` view,” notes Wood. “This will probably need to be tiled into smaller segments for delivery, and the right tiles to match the viewer’s head position need to be fed to the display. If it can be arranged so that the right tiles arrive without significant latency, the bit-rates could be ABR [Adaptive Bit-Rate] Internet-manageable – Megabits or tens of Megabits, but there are still studies to be made here.” 

Wood is open about the different types of display that could be applicable to VR applications. “VR can be something for very, very large screens that wrap around the viewer or group of viewers, it can be for HMDs [Head-Mounted Displays], and it can be for TVs, Tablets or smart phones,” he says. In the latter case, he suggests, viewers could change their ROI by moving the tablet, finger swiping, or by using a remote. “Allowing the system to serve these kind of devices in this way, as well as, and at the same time as, serving HMDs, will probably be a commercially important thing to do,” Wood believes.   

Asked if Augmented Reality (AR) is an application that could be included within the scope of his group’s research, Wood acknowledges that “we do need to take AR into account in our VR work. Among other reasons we could suspect that some HMDs will be designed to be used for both Augmented Reality display and Virtual Reality display. If you look at developments like the Microsoft Hololens, the AR looks amazing: AR via glasses at the same time as direct viewing with UHDTV could be a really terrific immersive experience.”

That said, Wood believes “the jury is still out on how the public will take to wearing the AR glasses, and on what content could be provided – and who could provide it.”      

Wood stresses that the DVB Project’s main interest is usually in developing specifications for a common delivery format: â€œWe are not in the business of developing production formats or display systems – they are for manufacturers themselves or other bodies. But we do need a delivery format that matches the production and displays used.”

The challenge for industry, he believes, will be to find a VR system “where the different pieces of the end-to-end jigsaw fit together – where there are no weak links in the whole chain. This is, if you like, the hard technical issue.”

Interested in virtual and augmented reality in television?

At Connected TV World Summit this week (London, March 16-17) there is an innovation workshop dedicated to the creative possibilities and technical requirements associated with VR, AR and also AI (artificial intelligence) in television. Session details are here.