Bring out a pair of those old red-cyan anaglyph glasses and take a look at the image above. Next, follow it up by looking at the image below:

Depth Ramps in Stereoscopic VR production:

The images are the start and end frames of a “Depth Ramp”. In conventional stereoscopic 3D film production, a Stereographer works first with the Director and Cinematographer to plot out the depth-script of how events (shots/scenes) will flow as the narrative progresses along.

Directors who understand the potential of what stereoscopic 3D can do for visual storytelling, can build entire “mood scapes” where the subtle interplay between depth budget and the scene being filmed can draw the audiences into the story. For instance – stereographers and Directors have been known to inflate the depth budget of a scene to emphasize happiness, as well as contract the depth budget on scenes depicting sadness or to heighten a feeling of claustrophobia.

Why and how is this relevant to Cinematic VR production?

Consider this:

Both Cinematic VR “games” or game engine driven Machinima – as well as stereo 3D live action VR capture can benefit from an understanding of the Depth volume (budget) in a scene.

For the first time, Visual story telling can take place on an Imax® like scale… where the screen is strapped to the audience’s face. The immersion is heightened.

There is more than a fair chance that if depth budget is not paid attention to in a Stereoscopic 3D VR production – harm can come to audiences, ranging from mild headaches to (sometimes) severe nausea.

What is a Depth Ramp?:

In a 3D movie, either conventional or Cinematic VR, the story usually flows sequentially and there will be a need to cut between scenes. Problems arise when the audience is viewing a scene of say – a first person Pov of an actor sitting at a desk watching a door at the far end of the room, waiting… followed by an immediate cut to a closeup of his hand pulling a gun from a drawer under his desk.

What happens is, the audiences eyes have converged (diverged actually) to ‘fuse’ the left-right images of the door at the far end of the room, and in the next shot they have to converge to fuse the left-right stereoscopic images of the closeup shot. It does not matter whether this is filmed as live action stereoscopic VR or created in a game engine.

To mitigate the possible hurt that will result in such a transition – stereographers usually employ a depth ramp – within the span of one second; sometimes less, sometimes more depending on the flow of the scene, the outgoing shot will be brought “closer” to the audience and the incoming shot (of the hand) will start off at a further distance. Things will finally settle in at the chosen depth for the next shot after this “depth ramp” has been executed.

Now, it will most likely be impossible to execute such a drastic cut, even with a depth ramp – that’s when a cat-in-the-window shot can/should be employed.



(photo credit: Riley)

Cat in the Window? – You know… That shot they have in movies that builds suspense or is used to cover-up a glitch when a bad take gets noticed only during post production? Well, that cat in the window shot – or clock-on-the-wall shot, when it’s located at mid-depth in the stereoscopic 3D scene, can help soften the depth transition and visual gymnastics an audience’s eyes have to perform when accommodating for the depth change from the door at the far end – to the closeup of the hand picking up the gun from the desk.

Cinematic VR depth Ramp: How is it done?

Described below are the steps for a live action shot, though a similar technique can be applied to a Game Engine rendered VR movie.

1) In Aftereffects apply the “offset” filter to either the left eye view or right eye view of the Stereoscopic Equirectangular footage.

2) Click the stopwatch icon to “animate” or keyframe the amount of H.I.T (horizontal image translation) that is required.

3) Over the span of about one second, slide the left image either to the left or right to move the “entire depth volume” of the current scene either in the Negative (out of screen) or Positive (into the sceen) direction, such that it will approximately match the depth of the next scene.

4) Rather than doing a simple depth ramp – it is advisable to meet the next scene halfway, i.e, apply a reverse depth-map to the next scene, so that it starts off half way closer to the depth of the current scene and then within a second settles in to the depth budget chosen for the remainder of the scene.

Taking an example from Dirrogate, the VR novel under production – the scene above is preceded by a flyby over a city, where the audience is floating over the rooftops of skyscrapers and then the camera ‘lands’ on the balcony showing a mid-closeup of the girl. The cut is abrupt and depending where the audience is looking (this is a 3D- 360 VR shot) we need to ensure the eyes have enough time to get accustomed to the environment.

We do this by starting the scene at a more comfortable depth – pushing the scene into positive Z space, and then transitioning toward the end of the scene, closer into negative Z space – to where the next shot will start – a closeup of a comic book page containing this scene as a panel, flipping by.

If red-cyan anaglyph glasses are not available, look at the red fringe on the center-left pillar move, as we manually step through the frames on the timeline in the sceeen-capture video, above.

(note: the current teaser available on GumRoad does not have a depth ramp applied, but will be updated)