by Tom Zerega
Founder/CEO, Magnetic 3D
(This is part three of a four-part series looking at 3D in digital signage. Read part one here, and part two here.)
5 key steps to achieving a better glasses-free 3D experience
1. First and foremost, consider a session on glasses-free 3D content and creative training
It will be a huge eye opener, a time-saver and overall a worthwhile investment. When creating content for a disruptive medium like glasses-free 3D that has lots of complex twists and turns, the biggest issue is and has been failed executions with a client who jumps in with both feet only to be let down by a final execution that wasn't ready for prime time.
For content developers it means having the discipline to avoid saying, "I can do this," and just rolling the dice, without really knowing how deep the rabbit hole goes. Creating glasses-free 3D content is not as simple as just creating content in a 3D program and exporting it to your favorite glasses-free 3D display. If that were the case, then everyone would be doing it. What makes it special is that you have to be trained in the art and science of developing glasses-free 3D content to really understand the physical limitations of each system, since they differ from one another.
It's also important to commit to a system and stick with it to learn how to truly deliver a quality glasses-free 3D experience on that platform time and time again. Developers also will benefit from coaching on glasses-free 3D creative direction that re-imagines content from a new perspective, the audience perspective. To be a pro takes practice to understand what works and doesn't in this new medium, and learning means making mistakes — but I don't recommend learning on the job, or rather on the client's dime as it sure to strain the relationship. Sure, you can figure it out a fair amount by yourself, but training is a great way to save yourself weeks or months of frustration by learning the tried-and-true tricks of the professionals and keep your clients coming back for more.
2. Avoid "window violations" at all costs
It seems like a simple concept, but it can be really hard to re-think your whole storyboard when objects popping off the monitor are no longer allowed to hit the frame of the display. Here's a tip: Think of the 3D display like a window, and you (the audience) are standing in your house looking out that window. The Challenge is this: Objects that come in the window must go back out of the window without hitting the frame, or they will be stuck inside. Hitting the bezel ruins the illusion of forward projection because the frame of the display (or the "bezel") is at a certain plane in real 3D space that your brain recognizes as being a certain distance away from you. So in our minds eye, that object can't be both popping off the monitor a foot closer to you than the actual display surface and then also hitting the bezel at the same time, because that object can't be in two places simultaneously.
The effect of a window violation attempting to pop content off the monitor toward the viewer causes the creative to go flat, or worse, it can be tough on the eyes and cause nausea as the viewer's brain tries to resolve the problem. There are some unique ways to do this without losing all your creativity, and there are exceptions to the rule depending on the type of object you are using — but to learn all that, I recommend you seriously consider suggestion No. 1.
3. Create a "Window-within-a-Window"
This concept shrinks the canvas size of your content so that you have a black border on one or more sides, thus leaving room for certain content to pop over the edge and appear as if it's outside the canvas. This effect is a surefire way to improve the illusion of forward projection off the monitor substantially further than it actually can go ("perceived pop") and also create a more gratifying glasses-free 3D experience overall. The idea is fairly straightforward, but it requires some creativity or the concept can become very stale. When objects are popping off the monitor, it's a great trick to have them eclipse some other content on the display that remains at screen level or even in the distance "in" the monitor to create nice depth cues.
Here's how the "Window-within-a-Window" effect works:
For example, say a "can" of soda is popping off the monitor and the background wallpaper says "Concessions," with only a few of the letters visible as some are blocked by the "can." The "Concessions" wallpaper is not the full size of the canvas and does not reach the edge of the monitor, but instead features a black border that surrounds the entire image. (A "window" of content has now been created inside the monitor bezel, which is also considered to be a window when dealing with 3D — hence the name "Window-within-a-Window.") The "can" crosses the border and eclipses the black area around the wallpaper as it comes forward. The effect will force the "can" to appear to be floating much further off the monitor than if it were not eclipsing the black border around the wallpaper. The Window-within-a-Window effect provides the audience with the "relative position" of the objects popping off the monitor from one another including the background.
4. How to deal with text on glasses-free 3D displays
One of the side effects of glasses-free 3D display technology is that the addition of the optical element lowers the resolution of the monitor, or the perception thereof. Depending on the glasses-free 3D display solution used and the steps taken by the content developer during the content creation process ahead of time, the reduction in resolution can be more than tolerable for the viewer.
Technically, the issue is that the optical element cuts across several pixels and shares sub-pixels to aid in creating the 3D effect. The problem with this method is that traditional 2D text doesn't look as crisp and clean through the lens for various reasons. A simple fix if you want to create plain text is to make the fonts bigger and avoid white text if possible. If you do have to use white, try to choose a font that is several pixels wide so that the sub-pixel sharing is less destructive, and also turn on anti-aliasing. To add contrast and sharpen the image even further, create a 3D version of the text and change the color on the side of the text to a darker shade than the front surface to add definition or shadowing – you'll find right away that the letters clear up, and it's easier to appreciate the 3D effect while decreasing the so-called ghosting effect (double image or blur).
With the 3D version of the text, you can push the content much further off the monitor than before and have it still maintain its clarity. Here's another tip: Avoid using After Effects (if possible), 2D-plus depth cheats or motion blur effects, as they will only decrease the quality that can be achieved.
5. Last but not least, conversion
My first words of advice: Don't take shortcuts. Over the past five years I've had lots of inquiries about converting 2D to 3D, but in my mind there should be a huge hazard sign telling you, "No! Turn back! Now!" or at least, "Enter at your own risk."
Why? Well, I hate to say not to do something without giving the reasons behind it, so here goes: The truth for one is that you are asking a computer to figure out where objects would be in 3D space — and computers don't do a particularly great job of filling in all the information that the human eye can instantly resolve.
The reason 2D-to-3D conversion does not work well, especially in autostereo, is that the information doesn't actually exist, and therefore the computer is estimating everything that it thinks should have been there at that point in time — or instead it stretches existing information to fill in the holes. With auto-stereo, artifacts as a result of the conversion are amplified by having to cheat the information for more than one view.
The second issue is that most people experiencing a glasses-free 3D display are not stationary, which would make the computer's job easier. Instead they are usually moving with respect to the position of the display.
The third issue is that most autostereo displays are too small for this process to be believable or to create enough pop for people to even notice that it's in 3D. The 2D-to-3D conversion in a movie theater works (when done well) because the screen is absolutely huge. It also works because the audience is not moving, so it's easier to fake the 3D effect because the amount of information to fill in is much less than if you were moving position and the computer needed to fill in other things behind the object popping of the screen (known as motion parallax). But with glasses-free 3D displays, especially for applications like digital signage, motion parallax is a key ingredient to creating the sense of realism that allows the brain to accept the illusion of 3D.
Motion parallax is what some refer to as the "look-around" effect, which means when you move one way, you can see around the object as if it were real, and the object doesn't follow you wherever you move. Well, 2D-to-3D conversion not only lacks motion parallax, but the computer tries to fill in pixels between the objects in the foreground and the background. And when it can't figure them out, it simply stretches the background and the foreground together and warps the image to fill the gap.
Unfortunately, algorithms that are available today fail to deliver the "WOW" experience that audiences are looking to see on a glasses-free 3D monitor. So my suggestion is to avoid showing 2D-to-3D conversion content. It always fails to deliver on expectations once people have seen real 3D images, and it's actually a disservice to the technology. Furthermore, it's a letdown for the client who so desperately wants that "WOW" factor, and ends up negatively impacting the industry as a would-be customer becomes disillusioned by the experience instead of charged up to do more.
The glasses-free 3D industry would benefit immensely from some self-discipline when it comes to content production. Display manufactures need to get involved and take an active role in the education of content developers to ensure they are trained in the art of creating quality glasses-free 3D content and not out there making it up as they go along. It's even more important in the industry's fragile early state that experiences fully captivate audiences and achieve highly desirable results for paying clients, rather than turning them off. The industry needs success stories. Bottom line, the stakes are extremely high, as anything less than brilliant content will result in a failed execution.
As with any new industry, it's not just that one project at risk, but the acceptance of the medium on the whole that could be called into question when the innovative, risk-taking pioneers get burned.
Zerega is the founder and CEO of Magnetic 3D, a glasses-free 3D display technology and media provider based in New York City. His roots are in broadcast, where he worked for nearly 20 years with major networks such as CBS, NBC and ABC/Disney prior to launching the Magnetic 3D brand and Enabl3D Technology in 2007.
Read more about 3D digital signage.