by Nikhil Balram
This is the question that came to mind for many attendees at the 2013 Consumer Electronics Show (CES) in January in Las Vegas. As usual, there were lots of interesting new consumer gadgets, with a special emphasis on networked and mobile ones. TVs were again a prominently featured item. But the main storylines for TVs were the connectivity and intelligence, and the step up to 4K resolution. A first-time visitor could easily think she was back in the pre-Avatar period of 2008 or earlier, when the big future themes were connectivity, smartness, and higher resolution.
The reality is that the second big coming, the “Renaissance” after the “Golden Age” of the 1950s, of stereoscopic 3-D (S3D) was probably overhyped in the euphoria around James Cameron’s Avatar. Other factors were the film industry’s need to raise ticket prices as an antidote to the shorter big screen life of movies, piracy, and other competitors for the consumer’s time, as well as the always-urgent need of the consumer-electronics market for “the next big thing” to drive sales.
One can reasonably argue that S3D has settled into a steady state in which it is one of the options that certain segments of cinema and home viewers appreciate and seek out, while others ignore. A very big difference from the 1950s is that S3D today rides on top of the overall trend of adoption of digital technologies in cinema – from production to display – so even a modest adoption by consumers can make the economics work. However, as technologists and vision scientists, we (the experts) recognize another very important factor at work here – the S3D that is available today is fundamentally limited and flawed, and these limitations and flaws play a large role in the diminished interest that consumers are showing.
The work of well-known vision-science researchers like Professor Martin Banks at UC Berkeley has shown very clearly and conclusively that stereoscopic 3-D has some basic limitations. The most fundamental and arguably most important one is the so-called vergence-accommodation conflict created by the presentation of stereoscopic 3-D on the flat, single-plane screens used in cinema and home today. The July 2008 issue of Information Display has two articles – “Consequences of Incorrect Focus Cues in Stereo Displays” by Banks et al. and “Scanned Voxel Displays” by Schowengerdt et al. that explain the issue and provide insights into possible solutions. This fundamental conflict is caused by the fact that presentation of stereoscopic images on a single plane results in an unnatural decoupling of vergence (the point at which a person’s eyes converge) and accommodation (the point at which the eyes focus), in contrast to real-world viewing where these two are always closely coupled. This conflict has been shown to cause viewer discomfort that manifests itself in different ways such as nausea, headaches, and tiredness.
The film industry has responded by moderating the amount of depth that is produced in S3D movies – the so-called “gentle 3-D”. In particular, this approach avoids putting objects of interest (i.e., objects the audience should be looking at) far behind or far in front of the screen. As a consequence, objects of interest are moved to the screen. This has reduced the possibility of viewer discomfort. But it has caused a different issue – a growing criticism from viewers that the 3-D effects are underwhelming and not worth the extra cost or inconvenience of wearing glasses. It is interesting to see letters to the editor in traditional home theater magazines in which the writers complain that they are used to mind-blowing audio effects when they go to the theater but that the visual 3-D effects seem to be much more muted – “Why can’t they blow my mind with the 3-D effects too – why are they holding back?” As experts, we know why “they are holding back” and there is no obvious solution. The key insight is that the current form of single-plane stereoscopic 3-D is just a transient and early phase of 3-D viewing.
So, in this special issue I have chosen to focus on the future instead of dwelling in the present or past. In my opinion, and that of many others, a much richer future awaits, when we can produce light-field or volumetric displays that can provide true depth, leading to rich and natural 3-D. At first glance, the technical challenges of producing full-blown volumetric displays seem to be so immense as to put them into a very distant future of interstellar travel and the kinds of “holodecks” seen on Star Trek. But there has been much interesting work done on compromises and subsets that could lead to a first generation of interesting and useful real 3-D displays. The work of Professor Banks and his colleagues Kurt Akeley, David Hoffman, and Gordon Love has shown that a surprisingly small number of depth planes are sufficient to produce the visual effect of continuous depth, suggesting that (depending on the application) perhaps even a display with 16 or fewer planes could suffice. The work of Dr. Schowengerdt and his colleagues has suggested that head-mounted or wearable displays could provide single-viewpoint volumetric displays that could have a number of uses.
The two 3-D articles in this special issue aim to look forward to that richer future of 3-D with a broader perspective of viewing and visual communication. The article titled “The Road Ahead to the Holodeck: Light-Field Imaging and Display” by Dr. James Larimer provides a good introduction to light fields in the context of human vision and imaging and display devices. It explains why light fields are the natural future of 3-D imaging and display and highlights some of the recent activities in this area. The paper titled “Communication through the Light Field: An Essay” by Dr. Stephen Ellis looks at the communication of information that occurs between display and human viewer. The article refers to the “ambient optic array” – the information that the human visual system detects from the light field. Dr. Ellis uses lessons from the past history of display systems to contemplate the major physical, economic, and social factors that need to be considered and addressed before the widespread adoption of light-field systems can become possible.
These two articles serve to remind us of the long and exciting path ahead for us as display researchers, engineers, marketers, and users. Indeed 3-D is dead. Long live 3-D. •