The 60+ page White Paper – “Glasses-free 3D cinema 70 years ago” – is an abridged extract from the second volume of “3D Displays and Spatial Interaction” a book that Barry G Blundell is in the process of writing. The document reviews early efforts (prior to 1950) to implement glasses-free 3D cinema. It focuses on work undertaken in Russia (S P Ivanov), Belgium (E Noaillon), France (F Savoye), and the UK (Dennis Gabor). Other pioneering work will be included elsewhere in the book. I had the chance to ask Barry a few questions about his fascinating “historic discoveries”
What is your background? I’m a physicist and engineer. As a university academic I’ve been able to undertake extensive theoretical and applied research into wide-ranging aspects of 3D systems and techniques. This background has also enabled me to publish five books in the field. I’m strongly committed to transdisciplinary R&D and particularly enjoy the dynamism and innovation that can be fostered by bringing together highly talented individuals from the traditional sciences and humanities to work on cutting-edge projects.
What brought you to 3D? My interests in 3D systems date back to my student days when I became frustrated by the challenge of formulating either on paper or in the mind meaningful models that included a third dimension. I don’t just mean models relating to potentially complex 3D problems like those frequently encountered in multivariate calculus and statistical physics (which may also incorporate a degree of animation), but even simple geometric models needed to solve basic trigonometry problems. Initially my research into 3D systems was very much technology-centric, driven by a desire to support the depiction of 3D images within a 3D space. However, as my interests in autostereoscopic 3D broadened, I soon realized the critical importance of also researching our remarkable sense of sight – the display and the visual system being two sides of the same coin!
Your free downloadable white paper (www.barrygblundell.com) gives a fascinating and, it seems, totally forgotten history lesson of glasses free 3D cinema. It seems like the Russians had the technology shortly before WWII and that Stalin provided early impetus to this work. How did it work and do you think the effects compare to today’s autostereoscopic 3D technologies?
A number of today’s autostereoscopic products are fundamentally based on the parallax barrier or lenticular panels. In the case of the former, a set of very fine opaque strips are arranged vertically in front of the display screen, whereas the lenticular approach uses a set of parallel cylindrical lenses. Both scenarios support glasses-free viewing. However in their classic forms of implementation all viewers must be located at an approximately fixed distance from the display. Obviously for 3D cinema and TV applications this doesn’t work.
Back in the mid 1920’s Edmond Noaillon was considering the implementation of glasses-free cinema and overcame the limitation of the parallax barrier by employing a set of non-parallel barrier strips. This so-called radial raster is arranged in front of the display screen with the opaque strips radiating out from a point located below the screen. He also devised a projection system in which the stereo frames were cast through the barrier onto the screen with reflected light passing again through the barrier before reaching the audience.
The key to the radial barrier is that it provides considerable freedom in viewing position and so allows the 3D content to be viewed from positions across the width and breadth of a cinema auditorium.
Researchers in Russia were the first to successfully apply Noaillon’s radial raster barrier to cinema and successfully developed appropriate 3D content. In the first half of 1941, glasses-free cinema flourished in Moscow – but came to an abrupt halt with Russia’s entry into WWII.
In 1947 glasses-free cinema re-opened and by this time the radial raster barrier had been superseded by the radial lenticular panel. A key weakness of the barrier approach is that light falling onto the opaque portions of the barrier is lost –the lenticular approach doesn’t suffer from this drawback and so supports the formation of significantly brighter images.
In parallel with the developments in Russian, Noaillon continued to work on the radial barrier. To eliminate image segmentation (caused by portions of the barrier blocking the passage of light) he introduced oscillatory movement of the barrier and also made use of multi-layer barrier structures which were intended to improve the form of ‘sweet spot’ viewing zones.
The effective use of oscillatory barrier motion is non-trivial and it wasn’t until the 1940’s that a practical system emerged in the form of the Cyclostereoscope – developed by Francois Savoye.
In comparing these early approaches with today’s autostereoscopic systems, it is important to bear in mind that the autostereoscopic pioneers were working with relatively primitive materials and projection systems, etc. The opaque strips employed in the implementation of the radial raster barrier used in Russia in 1941 were formed using 150km of copper wire and the weight of the overall barrier was ~6 tons! At that time, the production of the lenticular radial panel would have been an even more daunting undertaking.
It is also important to recognize that today’s large area autostereoscopic systems are expensive products – in contrast some of the earlier display paradigms could in principle (and with today’s know-how) be implemented for a fraction of the price.
In your work what is the most fascinating fact that you uncovered? Perhaps the fact that more than 70 years ago, numerous highly talented researchers recognized the importance of developing glasses-free, multi-viewer 3D solutions and were able to create innovative approaches – some of which could be realized using the materials and technologies available at that time.
Of course the success of the autostereoscopic technologies developed in Russia was only possible because they were supported by the development of techniques (both technical and artistic) needed to enable the creation of appropriate 3D film content.
What’s your view on modern 3D cinemas (with glasses) and what do you expect to see in the future? Although the glasses-based approach usually provides an excellent visual experience, I quickly become irritated by the glasses! This is very much a personal observation based on a belief that the glasses-based approach is a non-optimal solution. Without doubt we have the ways and means to deliver affordable glasses-free 3D for both cinema and TV.
Predicting in a few words the commercial future of 3D delivery is fraught with peril – especially as this is an area where disruptive technologies are likely to have a significant impact. However it is readily apparent that there is an urgent requirement for both cinema and TV to embrace glasses-free solutions in the near future. There is also a need to develop technologies that can seamlessly mix 2D and 3D content – allowing each modality to be used to maximum advantage. Given the diversity of 3D applications there will be a need for a range of display techniques – I doubt that any single ‘universal’ display solution will emerge.
In charting the future, much may be learned by taking a backward glance at the extensive and often highly innovative research carried out in this area during the last 150 years, and in using this to facilitate the production of technologies that satisfy our current requirements and work in harmony with, and capitalize on, our remarkable sense of vision.
Here is the introduction chapter of the free report:
In the following pages attention focuses on four exemplar strands of pioneering research carried out in the first half of the twentieth century in designing and implementing glasses-free (autostereoscopic) 3D cinema.
For nearly 100 years, it has been understood that stereoscopic techniques fundamentally based on conventional parallax barrier and lenticular methodologies are able to support multi-viewer glasses-free 3D – provided that all viewers are positioned at approximately the same distance from the screen. However, any approach that is to be successfully applied to cinema must clearly accommodate 3D viewing across the length and breadth of an auditorium. Central to the research efforts outlined in this document are innovative techniques which were intended to support this requirement.
The most rapid advances in early glasses-free 3D cinema and 3D cinematography in general occurred in Russia, and by 1941 Moscow cinema-goers were able to experience 3D on a screen measuring ~5m by 3m – without recourse to viewing glasses. Despite there being fewer than 400 seats, in a four month period (i.e. up until Russia’s entry into WWII), approximately 500,000 people took the opportunity to view 3D – glasses-free.
Three of the exemplar strands of research outlined here focus on work undertaken by Professor Edmond Noaillon in Belgium, Semyon Ivanov et al in Russia and François Savoye in France. In each case their glasses-free 3D cinema solutions utilised some form of ‘radial raster’ (in barrier and/or lenticular forms). Whilst Noaillon (the largely unrecognized inventor of this barrier geometry) focused on turning theory into practice via increasingly complex electromechanical techniques, Ivanov and co-workers adopted more pragmatic solutions. Further, shortly after the end of WWII they made the significant advance of replacing the radial barrier with a radial lenticular arrangement. As a result, autostereoscopic cinema soon flourished in a number of Russian cities. In parallel, following Savoye’s invention of the Cyclostereoscope, French audiences were quick to sample glasses-free 3D.
Although Dennis Gabor is widely recognized for his work in the late 1940’s concerning the invention of holography, his extensive and innovative efforts in developing viable forms of glasses-free 3D cinema have received little attention. This work forms the fourth strand of pioneering activity selected for inclusion here. Discussion is limited to aspects of two patents filed by Gabor in 1940 in which images recorded on lenticular film are projected onto multi-layer optical structures. Related patents filed by Gabor in the 1960’s will be discussed in greater depth in other sections of the second volume of ‘3D Displays and Spatial Interaction’.
There are many excellent publications devoted to the history of cinema. Early work undertaken in developing and deploying glasses-free 3D solutions has however received relatively little in-depth attention.
This document is intended to provide a technology-centric insight into several indicative approaches – these are considered in an accessible trans-disciplinary framework. Given the cyclic nature of 3D research coupled with the emergence of new materials, processes and technologies, an appreciation of past work can help in the identification of techniques which may be applied to the development of the diverse forms of 3D tableau needed to satisfy today’s increasingly complex visualization requirements. However in parallel and from the perspective of planning future investment, it is important to recognize that in the case of some applications (particularly those involving the visualization of complex data sets and/or high levels of interaction), support for binocular parallax based 3D is in itself only a partial solution – natural support for motion parallax may be of at least equal importance.