Ovum forecasts that VR video will generate revenues of $8.2 billion (€7.4bn) by 2020, which is impressive given that today consumer revenues around VR video are effectively zero – most content is either brand promotion, personal video, or technology demos. It also means that video will surpass video games as the number one revenue category within VR in 2019, as millions of consumers get portable mobile headsets and spend modest sums on content.
Despite this massive growth in revenues there is still a lot of confusion around what types of video are available and suitable for a VR treatment. Fundamentally, the following five types of video can be used in conjunction with VR headsets:
Normal SD/HD video in a VR environment. Essentially, this is the kind of service Netflix and Hulu have already announced for a variety of VR platforms, and is the killer app for “virtual cinema” headsets like the Avegant Glyph. When the user puts the headset on they can place themselves in a virtual environment (cinema, plush apartment, on the moon) with a flat “screen” in front of them on to which a video stream (SD or eventually HD) can be rendered.
Aside from some application architecture for VR controls and the environments, this is pretty much like streaming video to a tablet or smartphone – user authentication, frame rate, and catalogue are key.VR does offer some interesting future opportunities here, such as shared viewing with friends regardless of their location, and Minority Report–like navigation between video streams and services (tantalising for sports fans wanting to watch multiple games or fiddle with replays).
“3D film” in a VR environment. An extension of normal streaming video would be to use an existing 3D film source for viewing. This would not be as immersive as a true VR video, but more like an “embossed” version of a flat 2D film – or like one of those pop-up greeting cards – so you would not be able to move around the back of the scene even on rails. Given 3D film has largely failed in the home and been relegated to cinema experiences, this is never likely to be a major usage scenario – especially given how clumsy the 3D films will look in VR.
360-degree video. Opinions vary as to whether much of this should really be called VR video – while it is the single biggest application for mobile and promotional VR headsets and what most 360-degree cameras capture, most of these videos are not stereoscopic. This means you are effectively filming all around you (even above and below in some cases), but generating a standard 2D video. So watching these videos is like sitting in a fishbowl with video “pasted” on to the interior surfaces. It works great for landscapes and long shots, but falls apart as soon as objects/people are nearer as there is no depth perception.
Non-interactive/on-rails immersive VR video. The most common type of VR video will be on-rails experiences where a cameraperson has used a 360-degree stereoscopic camera to capture a scene. You will be able to “move” through the scene as it progresses, and look around and focus on different areas of the video, but you will not be able to interact in any way with its contents. Most of the early VR documentaries and shorts work like this.
Alternatively, users may be presented with multiple locations that thry can “hop” between to view an event (for example, track-side, bleachers, and sky-cam in a sports arena).
Fully immersive VR video. To truly utilise the VR platform, video should be fully interactive, where you can walk around in a scene as you please – more like a play or art installation than film. So far this is not really possible from a technical perspective, as the players would have to be rendered on the fly rather than pre-packaged. Computation photography offers hope here, but is still largely a theoretical space with a few pioneers like Lytro and more recently
Intel leading the charge. This also raises questions about the structure of entertainment – how can you successfully advance a story if you are not sure the viewer is looking at the right thing. This form of VR video is getting to the point where it will be indistinguishable from VR gaming/interactive experiences.
Regardless of the type of video one produces for VR, new challenges will be introduced around editing, storytelling, and duration.