Micro Project: 3D Storyboarding?

This micro project is a follow on from my previous blog-post about the usage of 3D in production illustration for The Hobbit. As promised, here I attempted to apply 3D anaglyph rendering to the process of Storyboarding. 

As said before, not being a fan of 3D in its form in Cinema, I was even more clueless at how you even create the most basic image through this 'technology'. Below are my crude- and decidedly failed- attempts at creating a basic 3D image.

In Photoshop, a quick layered image of a Tree. The tree and the grassy plain beneath are the foreground; the mountains/hills are the middle-ground; and the sky becomes the background. This became the template image.

In researching anaglyphic imagery, the most common colours used are Cyan and Red. Below was a rendering of the same image in a near cyan colour.

And the other in red.

The last stage was to layer the colour versions over the original black and white image, separating- as I had assumed- both the cyan and red layers appropriately to create the 3D effect. If you have any 3D glasses on...you will see how I failed miserably. But this was an important stage in realizing I had to explore a more economical and less convolated means of achieving this. And more importantly, achieving this successfully.

Second Attempt. 

After some re-thinking and tinkering with layers, I discovered that, lo and behold, Photoshop had already done most of the leg-work for me. Each image on a layer has colour channels. These colour channels are divided into Red, Green and Blue. (the latter two more or less become Cyan.) With a little channel/layer movement, I finally created my first anaglyph 3D frame.

The process became simple:

• Create your image
• Duplicate layer
• Click colour channels
• Select Red layer
• Select entire canvas (ctrl+A)
• Select move tool
• Move Red layer to the left as desired

With that figured out, it became clear I needed to consider how this could be used in a Storyboarding context. Simply rendering an image in 3D does nothing to the narrative or logistics of boarding. So the question was: what could 3D solve that we can't already? 

Although 3D is a misnomer of sorts- in that we are creating the illusion of three dimensions via flat drawings anyway- I came to the conclusion that the most obvious aspect that 3D could effect is the visual description of Lens types. As any Storyboard artist knows, there is already a set means of visually describing camera movements and lens types through our own 'language' of arrows and icons.

Directors, cinematographers, and illustrators know what these arrows denote. Below are examples (from my Brief planning) of how I would communicate a Dolly Zoom, depth of Field, jump cuts, etc.

Now I needed to see if 3D could recreate this effect. Remember, the point was not that this sequence would become a 3D feature; but rather the 3D element here has become a tool to describe a camera movement, even if the final sequence remains 2D in post-production.

Whilst creating this image, I kept in mind the need to be able to separate these into layers. The hand becomes the foreground (untouched by the anaglyph rendering) whilst the rest is 'pushed' back.

A second iteration fine tuned the layering process. This time I was able to control the amount with which the separate layers were pushed back. The hand remains untouched, whilst the head's red layer is only moved a fraction. The background is pushed away significantly. I think this is somewhat a success in that it certainly re-creates the result of a dolly zoom.

Here was another test in exploring 3D's enhanced visualization of field depth. Below was a sketch showing how I proposed to breakdown one of my storyboard frames (from the Brief) into a clear division.

Again, this was the first step of turning the entire image into a 3D image.

And here the clear three-part separation can be seen.

Jump cuts are often a tool employed by Storyboard artists, or tracking shots that span a single frame with inner-frames denoting what the audience will see, and what they won't.

Repeating the same process I isolated an area with which I wished to turn 3D. In this case it became a recreation of a jump cut. 

A second alternative, and perhaps a little easier to read. With the previous attempt, the 3D became difficult to read as it pushed the jump cut away from us. This time I switched it around, pushing the background away, and leaving the area with which we are supposed to be observing intact. 

Evaluation.

After these experiments, I've struggled to justify an alternative means of visually communicating camera motions commonly found in storyboards. Whilst the dolly zoom and jump cut remain effective, the time and process is a negative that cannot be ignored. Moreover, the current system of arrow icons as employed already successfully communicate these logistics. 

Granted, I was only able to use the anaglyph approach. There are other forms with which this could've been easier; other technologies that are not available to me. That said, by simply evaluating the benefits of turning a storyboard into 3D, we can see that 3D isn't solving any problems that are already present in Storyboarding. In truth, they simply add to the execution time for the artist, where a simple arrow would communicate all of the above faster, easier, and just as cleanly.

Here's a breakdown of the pro's and cons.

Although I am somewhat pleased to see that this doesn't work- as I've said before, I'm no fan of this process in Cinema-it is interesting to see that where The Hobbit has discovered a new means of applying 3D thinking to the creative process, it may be that I've hit a brick wall where it ends. There's nothing productive to be found here at the Storyboarding stage. In the animatic stage? Certainly. In Production illustration? Maybe.