GB2401953A - Stereoscopic display device and method of creating pseudostereoscopic moving images - Google Patents

Abstract

A method and apparatus for producing three-dimensional images from two-dimensional motion picture film, videotape, television or other moving image material, is achieved by selecting for transformation that original moving image material which contains at least one of nine defined categories of motion parallax exhibited within the frames. The original material is re-displayed on a pair of displays or a wide-screen television monitor (1) with time-delayed copies displayed to the right (2) and left (3). The screen is viewed through a wedge prism pair (4 and 5) whereby the viewer perceives three-dimensional depth in the moving images. Also disclosed is device 4 which can transform a pair of two-dimensional X-Ray images on adjacent television monitors into a three-dimensional composite image when viewed as described above.

Description

STEREOSCOPIC DISPLAY DEVICE AND METHOD OF CREATING

PSEUDOSTEREOSCOPIC MOVING IMAGES

Field of the Invention

This invention relates to a stereoscopic display device, and to a method of creating pseudostereoscopic moving images from 2-dimensional moving images.

Background to the Invention

A Stereoscope designed for two drawings was used as early as 1832, one making use of two photographs was constructed in 1841, and the first stereoscopic camera appeared in 1853. These developments closely followed the invention of still photography and there followed an intermittent but sustained search for visual systems which replicated the condition of three-dimensonal vision found in the animal kingdom in an attempt to improve on the limitations of the still photographic image. With the advent of cinema, stereoscopic cinematography and projection mechanisms were developed in the 20th century.

All such systems Involved recording devices designed with the intention to produce a three-dimensional image when appropriately viewed. However the history of the motion picture film and television industries through the 20th century saw two dimensonal motion picture film and, subsequently, television become the normal mode of production and viewing, and threedimensional motion picture film a specialist minority activity, primarily because of the considerable extra costs involved. The majority of the voluminous products of 20' century two-dimensional motion picture film, videotape and television production in various genres now reside in archives. Their viewing destiny has always been assumed to be in the twodimensional form for which they were initially intended.

While two-dimensional film has been notably successful as an entertainment and educational medium, it has always been limited by the inherent restrictions of the single photographic lens. Two-dmensional photographic images are interpreted by the human eye and brain as acceptable representations of reality. They depart however from what the brain normally "sees" via the eyes. The human eyes are separate receptors, some two and a half to three inches apart, and the two eyes consequently see a slightly, but significantly, different image. The brain Integrates the left and right eye Images Into a - 2 single three-dimensional composite. The human brain, and the brains of many animal groups, evolved In a stereoscopic viewing environment. The human brain can obviously tolerate the limitations of the two-dimensonal photographic Image, and the successes of the still photography, motion picture and television industries bear witness to this, but it finds three-dmensional representations far more dynamic, in fact, more realistic. This fact has fired the inventions leading to three-dimensional representations of the world which can be summarized, in the photographic context, as methods for directing two different images to a viewer's left and right eyes. A majority of schemes involved two cameras placed side by side producing an image pair similar to that on the two human retinas. In normal living the human brain extracts information about the apparent differential distance of the observed objects within the field of view. The left eye sees closer objects displaced to the right while these objects are seen by the right eye displaced to the left. More distant objects are likewise displaced but less so, and beyond a certain distance there Is no significant displacement. This set of differential displacements Is what the brain appears to compute to achieve normal human three dimensional vision. In consequence, when a pair of images exhibiting such parallax is presented, one to the left eye and the other to the right, the composite result appears three-dmensonal. The visual cortex of the brain combines the two images, by a process called stereopsis, into a three-dimensonal image of the field of view.

Putting aside holographic systems employing lasers, the prior art of three dimensional viewing systems Includes, in approximate order of disclosure: A. Hand-held viewers of various sorts, all transmitting a slightly different image to the left and right eye; B. Polarizing spectacles; C. Coloured spectacles (called the analglyphic 3-D process); D. Lenticular systems (employing autostereoscopy) (requiring no viewers as in A., B., and C.); E. Shutter system spectacles; F. Random dot stereograms; and G. Flat screens displaying vertcal-blocking processed images (requiring no viewers as in A., B., C., and D.).

A useful appraisal of the advantages and disadvantages of these various approaches is made by KLEINBERGER et al, WO0118589. Of the above, four, namely: B. polarizing spectacles, C. coloured (red / green) spectacles, E. shutter-system spectacles and G. flat screen blocking, have been developed or employed for moving image stereography.

Obvious minor or major disadvantages, inherent in these four, are, for example: B. Polarizing spectacles involve light loss due to inherent absorption; C. Coloured spectacles alter the transmitted wavelengths preventing their use with colour film, videotape and television; E. Shutter-system spectacles increase the perceived flicker on the viewed screen; and G. Flat screen blocking results in increased grain and lowered resolution.

Earlier systems for recording stereo pairs on motion picture film employed a camera with twin lenses systems and paired projectors and polarizing or coloured spectacles. These proved too expensive, involving as they did twice as much film stock as normal, difficulties in achieving technical parity of image characteristics, and complicated duplicate synchronous projector requirements Attempts to achieve a three- dimensional moving image illusion using a single camera, be it recording motion picture f Im or videotape, have centred on the phenomenon of motion parallax.

Depth perception of a three-dimensional scene results from various kinds of information, viz.

a) learned cues: cultural familiarity with what is being viewed b) monocular cues: relative size, shape, textural gradients, linear perspective, occlusion, interposition, differential light and shade, colour, differential focus, position relative to horizon and motion parallax. Motion parallax Is perceived relative motion of near and far objects when either the observer or the object moves laterally and is arguably the most important monocular depth cue.

c) binocular cues: retinal image size, accommodation, convergence and retinal disparity.

Retinal disparity (also known as "binocular disparity"), arguably the most important binocular depth cue, refers to the amount of difference between the views afforded by the two eyes. The disparity provides cues by which an object's distance from the eyes is assessed. Disparity is inversely proportional to the distance of an object to the eyes - the closer the object is, the greater is the disparity. It Is the degree of change between an object and the background, as the head moves in space, that provides continuing cues for assessing the distance between an object and the background.

Mobon parallax and retinal disparity operate to sustain the threedimensional Impression we have of the world, and it is primarily these cues which make it possible to optimistically approach the conversion of two-dimensional to three-dimensional moving images.

The present application for a patent concerns a technique and system which makes it possible to meaningfully transform much existing traditional motion picture film, videotape, television and X-ray images into a form which produces a three-dimensional viewing experience. Most Importantly, the viewer will be looking at exact and unaltered copies of the original two-dimensional film, videotape, television and X-ray images in a new manner which gives the convincing impression that the image is three-dimensional, such that the image presented to the left and right eyes of the viewer is appropriately different. In this way, for example, certain archive motion picture film from, say WW2, can to be viewed, for the first time, In three dimensions, exactly as the originating cameraperson would have seen the scene being filmed. It will be seen below that a significant proportion of the billions of feet of existing motion picture film, videotape and television material is amenable to three-dimensional conversion.

Until now it has been generally presumed that a very restricted type of motion picture or videotape material, namely a precise "tracking" shot, where the camera moves In a straight line, may produce footage, certain individual adjacent frames of which will have been recorded from a position effectively equivalent to the distance between the human eyes. Such footage, it was correctly assumed, could be employed to produce a three-dimensonal image, given special processing, projector, or viewing treatment.

However, the present disclosure takes our knowledge further.

Disclosures in this area include: - 5

US4925294 (1990-05-15) Gershwind & Handal wherein some degree of threedimensional depth is achieved by a human operator assigning depth information to various elements of the image combined with computer analysis. The high degree of human intervention necessary makes this approach applicable only to small volumes of material and full threedimensionality of the transformed material is not claimed.

W09513564 (1995) White wherein motion picture footage is projected by a non-standard, hence expensive, twin-lensed motion picture projector and viewed with polarized spectacles, or digitised and assigned to a RAM frame buffer in a computer and displayed as superimposed combined polarized images for viewing with polarized spectacles involving inherent diminished transmission. Another embodiment outputs to expensive head mounted liquid crystal displays. White's static camera and moving elements within the frame, which produce motion parallax, prefigure Category 5 below.

US5953054 (1999) Mercier refers to the earlier US4807024 (1989) McLaurin et al (University of South Carolina) wherein a system is described for generating motion picture film or videotape images, of the earth surface beneath, from a constantly moving airplane flying in a straight line. Pairs of stereoscopically related images of the resultant footage, selected by one of the pair being time-delayed, are presented alternatively at an average frequency of 10 frames per second. The viewer perceived a three dmensionality without the use of special spectacles, and the invention could be seen as having particular value in aerial reconnaissance.

Mercier, developing this earlier work, the paired images either alternated or superimposed and viewed with electronically synchronized shutter lenses or polarized lenses, the disadvantages of these having been discussed already. Mercier stresses the linearity of camera movement and the need for the camera axis to be perpendicular to that of the camera's movement. This prescription, to be discussed shortly, is echoed In US5963303 1999 (and subsequently EP1059559 2000) Allen.

Allen describes a system In which a still, motion picture or videotape camera is moved In, again, a straight line (a "tracking" or "trucking" movement). Again a time- delay procedure is applied to one of the duplicate resulting images, or sets of frames, to produce a stereo pair and a stereoscopic effect when viewed. Allen however describes no specific process whereby the stereo pairs may be constructed and viewed. He refers only to the job of the film or video editor in this context. However he provides detailed useful calculations for computing the time-delay or frame-separation required to achieve a correct stereo pair. Employing various examples of footage shot in different environments his calculations relate the many absolute technical variables and estimates involved. His examples do not depart from a camera moving in an implied straight line (e.g. "an airplane is travelling at 240mph", and "old film footage, a boat motoring up the Amazon river was shooting a motion picture camera of the villages along the river" and "photographs taken from a World War II aircraft flying over the remains of Berlin").

However, some of his formulae involve estimates of the distance of objects from the camera lens In existing motion picture film, which are well known to be difficult to make, even for film professionals. Also Allen states that motion picture film cameras conventionally run at 32 fps, which is not so - it is 24fps. Nevertheless his distinction that new or existing "tracking" footage is amenable to the extraction of stereo pairs is correct, If far too limited, and his work, following Mercier and McLauren et al, prefigures Category 1 below.

US6392689 (2002) Dolgoff adds valuable observations, demonstrations and hypotheses concerning the role of the eye's convergence and accommodation, and how the brain computes such information in producing a threedmensional interpretation.

His display system, employing electronic delay, polarization and two or more video projectors offset slightly, and means for motion detection are however located in the arena of illusion-production rather than that of the present application. Dolgoff's " delay of one field (equal to approximately 0.0167 seconds in N.T.S.C. television) to about 0.0245 seconds, is preferable, depending on the motion speeds within the movie.") does not match the findings In the present application, and also it is important to note that, for example, high-speed cinematography imposes a further variable in the calculations of appropriate time-delay or frame displacement. His reminder, that all related attempts to extract threedimensional effects from original two-dimensional moving image material produces pseudostereoscopy, is noteworthy.

Summary of the Invention

According to the invention, a stereoscopic display system for moving images, comprising a pair of display screens arranged side by side, means for displaying on the screens respective ones of a pair of moving stereoscopic images, and at least one viewing device positioned in front of said screens and comprising optical means for each of a pair of eyes, constructed and arranged such that each eye of a person looking through the viewing device will see a respective one of said pair of moving images and the person will perceive a moving stereoscopic image.

The optical means may comprise a prism, the pair of prisms being suitably angled to enable the viewing person's brain to combine the images in each pair to produce the stereoscopic or three-dimensional perception of the moving images. Alternatively the optical means may comprise at least one lens.

The screens may be light-emitting displays, or regions of a single lghtemitting display. The or each light-emtting display may be of any type; suitable types include cathode ray tubes, liquid crystal displays, and light-emitting diode displays. The light emitting display or displays could simply be as a result of back projection on to a translucent screen.

Alternatively, the display screens can be light-reflecting, and the means for displaying may include a pair of projectors. These are suitably video projectors, but could also be film projectors, although synchronization of these might be more difficult than with video projectors.

A plurality of viewers may be arranged in front of the screens, whereby a plurality of people can view the moving images at the same time.

The invention also provides a method of creating pseudostereoscopic moving images from an original sequence of 2-dimensional recorded images, comprising creating a sequence of pairs of images in which the first images of the pairs comprise the original sequence of images, and the other images of the pairs comprise Images from the original sequence time displaced relative to the first images.

Preferably, the original sequence of Images results from or represents non-linear movement of the recording camera.

The Invention further provides a method of X-ray examination of an article, comprising recording a first X-ray image of the article using an X-ray Imaging apparatus, displacing the article relative to the imaging apparatus and recording a second X-ray Image, displaying the pair of images side by side on a display screen or screens, and - 8 vewing the Images using a viewing device positioned in front of said screen or screens and comprising optical means for each of a pair of eyes, constructed and arranged such that each eye of a person looking through the viewing device will see a respective one of said pair of images and the person will perceive a stereoscopic image.

The present invention however makes possible the viewing of many types of motion picture and videotape footage in three dimensions, and is not restricted to a precise "tracking shot" (Mercier, McLauren et al, and Allen), or the static camera shot (White). Indeed, the mobon parallax characteristics, which make possible the three dimensional effect, are not only produced by linear camera movement, but occur when any object, within the scene, moves relative to any other and to the camera. In consequence, two-dimensonal motion picture and videotape footage in which fictional characters move within the frame, people, animals or objects In documentary footage move within the frame, where the small camera movements intrinsic to "hand held" camera footage, and other categories of relative movement between the objects viewed and the camera lens, all conspire to produce a three- dimensonal interpretation in the viewer as a result of the procedures of this Invention.

It is possible that all previous workers, excepting Allen and, possibly Dolgoff, have subscribed to a notion of the primacy of the precise "tracking" shot, as that containing stereoscopic potential' because of their belief In the precision of the definition of stereographic images. So-called normal stereographic images are those recorded at positions equal to the nter-ocular distance, but, of course, the separation of human eyes has a mean of some 2.5" to 3" (63.5mm to 76.2mm) but quite a high standard deviation (the latter reflecting age, gender, and ethnicity). So normality is only a statistical fact.

What is normal for one human may not be so for another million. So-called hypo stereographic images are those recorded at distances less than the distance between the viewer's eyes - but for another viewer they could be normal. So-called hyper stereographic images, those recorded at distances further apart than the distance between the viewer's eyes, may be normal for another viewer. This does not detract from the classificatory and descriptive usefulness of the three terms, but the continuum of stereographic potential In image pairs needs to be remembered. What may be an actual normal stereographic image for one viewer can be an effective hypo- or hyper- stereographic image for others.

Further to this, in the search for methods to extract three-dimensional information from two-dimensional motion picture and videotape material, it is useful to discriminate between actual mobon parallax producing a stereoscopic condition, and effective motion parallax likewise producing a stereoscopic effect. Actual motion parallax, as has been said, is perceived relative motion of near and far objects when either the observer or the object moves laterally and is arguably the most important monocular cue. This common dictionary definition has perhaps made workers focus on the "tracking" shot as necessary for stereoscopic transformation, as it is the most obvious camera articulation producing parallax. But other types of shot, (and Allen indicates the static camera shot) and other types of film content produce effective motion parallax and resultant stereo pairs.

Before classifying the types of material which contain actual stereoscopic potential, it is important to recognize an intrinsic characteristic of human vision which cinema and television professionals, particularly directors and editors, have learnt to employ. The human eye has a very narrow area of focus, which is notionally one degree, where the light rays are focussed on the highly sensitive spot on the retina called the macula. Although we are generally unaware of it, most of our eyes' field of view is out of focus, all the time. Only a small circular area of the field is in focus, on the macula. Our eyes unconsciously scan the field of view by rapid eye-movements, called saccades, each static eye position presenting a small area of focus and a wide zone progressively unfocussed. The precise plane of focus is achieved by the eye's so-called accommodation and convergence, in which, respectively, the lenses are reshaped by the eye's lens muscles, while the ocular muscles direct the whole eye. (Arguably this saccadic propensity conferred such a survival advantage that it became an inherited trait in early mammalian development.) This phenomenon of saccadic eye-movement occurs while we view motion picture films and television, and filmmakers and editors have come to understand how to control or utilize these eye- movements. When presented with a static image containing a moving element, the human eye will be attracted to, and follow, the moving element. - 1 0

The success of animated cartoons relies in no small part on this fact. Given that the majority of the screen is out of focus for the cinema audience, and that their eyes are unconsciously drawn to elements that move, it follows that there are significant categories of motion parallax in parts of the frame that can produce an effective three dimensional impression - a metapseudostereoscopy, extending Dolgoff's correct reminder that all manipulations in this area produce pseudostereoscopy.

finally, it is worth ponting-out that because gravity, an invisible and effectively vertical force, results in the condition that the vast majority of visible movements on the earth's surface are effectively horizontal, and because the human eyes are displaced horizontally, motion parallax will be produced by the majority of such lateral movements in our everyday lives and while watching the screen.

Categories of motion picture, videotape or television two-dimensional shot which yield a convincing three-dimensional transformation are: Category 1 (exemplified In part by Mercier, McLauren et al, and Allen) Straight line "Tracking" shots with the camera angled (pointed) through any of the 360 horizontal degrees, be the camera on a smoothly running "dolly" or vehicle or hand held. (In the case of shots "straight ahead" or "straight behind", and depending upon the focal length of the lens employed, a small central area of the screen will exhibit virtually no mobon parallax, but the majority of the surrounding screen, and increasingly so towards the edge of the frame, there is exhibited a motion parallax such that three dimensonal transformation is possible. And it is this that the eye will be attracted towards, by unconscious saccades, and focus on, when necessary, by accommodation, on the screen.) Category 2 Curved and circular "tracking" shots, perhaps because a circle can be constructed from an infinite number of tangents and, as a result, contiguous frames closely resemble those recorded had the "track" been straight.

Category 3 Handheld shots, with stabc objects in the frame. In these, the intrinsic body movements of the cameraperson impose motion parallax upon the scene being filmed.

Category 4 Handheld shots with static and moving objects within the frame, whether moving laterally, towards or away from the camera. This is one of the most - 1 1 common categories of "newsreel" footage. It characterized almost all the combat footage of WWII shot by military cameramen whether Allied or Axis or Japanese.

Category 5 (exemplified by White) Static shots, with no camera movement, but in which one or more objects within the frame move laterally. (A few exceptions to the general lateral movement include rain, when there is no wind, and express trains filmed from bridges directly above.) Category 6 Panning shots (where the camera is rotated horizontally) and some objects move in the same or the opposite direction as that of the pan.

Category 7 Crane shots (where the camera rises or descends) in all such movements apart from exactly vertical. (In exactly vertical shots, extremely unusual as they are, and with no lateral movement of any object in view, no stereoscopic effect will be achievable.) Category 8 Zoom shots (where the focal length of the lens is varied) but only In combination with at least one of Categories 1 to 7. (Because a zoom lens only varies the focal length of the lens and hence the field of view, the relative location of objects in the frame is unchanged and no motion parallax is produced by the "zooming in" or the "zooming out" articulation alone.) Category 9 Any combination of the above Categories 1 to 8. (It is worth noting that the art of cinematography has steadily developed and one feature of this progress is that shots increasingly, and particularly since WW2, contain more than one articulation, e.g. a "tracking shot" Incorporating a "pan" and a "zoom" is common.

Examples from existing archive two-dimensional footage transformed to three dmensional footage: a) Black and white film. A camera, located on a moving vehicle, pointed some 60 to the right of the direction of travel, moves steadily through the streets of Berlin In 1945. The shattered remains of buildings rapidly pass to the right, gaping windows revealing more gaping windows beyond. The buildings fall away revealing a wide road to the right and an extended vista of desolation. Scattered pedestrians walk at various distances from the moving vehicle. A dog runs by. The road divides into two parallel curved routes, and an automobile enters the frame, moving left across the screen, overtaking the camera-vehicle. Foreground rubble, pedestrians, - 1 2 automobile, passing shattered buildings beyond - all sculpted In three- dimensional relief. (see: Categories 1 and 2) b) Black and white film. The camera is fixed. A shipyard in the bright sunlight of mid day. To the left, a large WWII merchant vessel lies at the dockside, its bulk placed obliquely to the camera's direction. It is motionless. However, three elements of the shot move. In the centre of the screen a large army tank, suspended on wire hawsers from the arm of a crane, swings slowly left, towards the deck of the ship.

As it swings it rotates slightly. The metal frame of the turning crane to the right rotates slightly. A man walks to the left, apparently directly beneath the huge tank, and seemingly unaware of its presence above him. Some distance beyond the tank and the man, smoke very slowly rises from a ship's funnel. In the original two dimensional shot, the man appears to be directly beneath the swinging tank and the rise of the smoke is hardly discernible. In its three-dimensonal manifestation, the man is now clearly beyond the sharply three-dimensional tank, and the smoke is strongly sculpted. Our eyes saccade to the tank, then the man, and then the smoke. We do not notice that the ship's structure has no three- dimensionality, because of its stasis, and because the monocular cues of perspective and size give it apparent depth. A pedantic categorization would exclude this shot from being actually three dimensional, but a filmmaker, knowing how human eyes watch shots, would include it, because of its (very) effective three-dmensionalty. (see: Category 5) c) Black and white film. The camera may be hand-held or on an uneven wheeled support. It is pointed at some 30 to the left of the direction of forward movement. The image is of a Jewish ghetto street market, probably late in the 1930's, somewhere In Europe. The camera proceeds forward along the street passing many people buying, selling, inspecting goods for sale in open stalls. All the various gestures, postures and movements of the people, the buyers, the lookers, the sellers, and others seated, waiting, one man stooped, sifting through things in a box, the cacophony of objects on display, are all seen In rich three-dimensional relief. (see: Categories 1, 3, and 4) - 1 3 d) Black andwhite film. A hand-held, rather shaky shot. An island somewhere in the Pacific, presumably 1945. A Japanese military man of high rank is slowly progressing along a small line of airmen about to leave for a kamikaze attack on the US fleet. The men, in a line oblique to the camera's direction, are not at attention and make small independent movements. The wind makes items of clothing flap.

There is a gap in the line of men, and beyond, a little lower down, a man has his back to the rest. He is beside a small pool fringed by reeds moving in the breeze.

As he looks back over his shoulder towards the men, the ceremony and the camera, his movements reveal that he is urinating Into the water. His movements have drawn our eyes away from the ceremony and towards his image, clearly further away than the line of men. All the players in this scene are cast in the clear relief of the three-dimensional image.

e) Colour film. The camera is steady and hand-held. The subject is a man, his head and shoulders fill the frame. He is wearing a greenish uniform and a military hat.

The camera is somewhat below him, looking up at him. Beyond him, the side of a building with windows. He is looking out over the camera somewhat to the right.

He looks serious and appears to be thinking. His frame moves slowly to the right and then the left, his head turning slightly as though looking for something, and the camera follows his movements, keeping him more or less centred. The movements of the man are cast Into strong three- dimensonal depth, with the building, well beyond, and aspects of his head are revealed in depth as he turns slightly. It is Adolph Hitler at the Berchtesgaden, his mountaintop retreat amid the Bavarian Alps.

Because a significant proportion of existing motion picture film and videotape footage exhibits at least some of the characteristics outlined In Categories 1 to 9, a vast amount of existing material stored In archives, museums, film studios, and private homes could be transformed to a three dimensional form. This will not only provide an enhanced public attraction in, for example, museums. It will also provide film researchers with more diagnostic cues to interpret meanings in archive footage.

Three-dimensional motion picture film has mostly been produced as "spectacle" for a popular audience and consequently mostly dismissed by critical minds. Indeed, the - 1 4 peak of three-dimensional motion picture production was in the 1 950's when generally sub-standard scripts were shot In 3-D as part of the movie industry's attempt to compete with the growth of domestic television, which was taking away its audiences.

The present invention is in not primarily aimed at producing a "spectacle", but rather a powerful tool for museums and other archival media-users to present, explore and research existing motion picture and videotape footage, and, in some ways to bring it alive, given the human brain's natural attraction to three-dmensional images.

An important feature of this invention, distinguishing it from anything in the relevant prior art, is the fact that the viewer of the result of this Invention can automatically and easily confirm the veracity of the three-dimensional re-presentation of the original footage. It is crucially important for museums, archives and historical researchers, among others, to be completely confident that they are looking at unmediated footage, uncut and an accurate and exact record of what was originally recorded on film. Now that computer generated synthetic motion picture footage has become a standard means of producing images which appear to the viewer to be authentic, it is of even greater concern to many museum, archival, research, educational and other users that original material remains "undoctored". This invention guarantees and foregrounds this, because the viewer is looking unarguably at the original material.

This invention thus releases the stereoscopic potential that lies presently untapped In the majority of shots involving movement both of the camera and of the subject (Categories 1 to 9 above) In material held in motion picture and videotape archives. It is not limited to material produced by camera tracking movements, but the stereoscopic potential is equally revealed by most types of movement of elements within the frame, be they that of actors, documentary or news Images of people, animals, plants or objects.

This invention transforms the original material into a profoundly more engaging form, and, of course new film, videotape or television material may be shot with this invention in mind for the post-production stage.

The most common name for the product of the motion picture industry is the term "movies". Movement became a defining characteristic of motion picture films very early on, and has remained so until the present. Television likewise has employed those features of the "movies" which engage the audience. It is clear therefore that the motion - 1 5 pcture and television archives of the world hold voluminous amounts of footage which are amenable to being transformed into a three-dimensional experience for the viewer as a result of this invention.

Brief Description of the Drawings

In the drawings, which illustrate exemplary embodiments of the invention: Figure 1 is a diagram, not to scale, representing the "individual" embodiment; Figure 2 also not to scale, illustrates a possible mounting for the pair of wedge prisms; Figure 3 is a diagrammatic plan, not to scale, showing an example of the "cinema" embodiment; and Figure 4 is a diagram, not to scale, representing a conventional X-ray security scanning apparatus.

Detailed Description of the Illustrated Embodiments Referring first to Figures 1 and 2, a wide screen television monitor 1 has two frames displayed comprising a stereo-pair, the right-hand frame 2 and the lefthand frame 3. The human right eye 6 and left eye 7 are viewing the television monitor 1 through, respectively, a right eye wedge prism 4 and a left eye wedge prism 5. A mask 8 with two rectangular holes 10, for the two eyelnes 9, is located in front of the wedge prisms, such that the viewer sees only the stereo-pair 2 and 3 displayed on the television monitor 1, and peripheral vision is obliterated. The dimensions of the rectangular holes are 4:3, the general motion picture and videotape and television proportion, but alternative, or variable masks for different frame dimensions, according to the characteristics of the original material, can be put in place instead. The pair of wedge prisms 4 and 5 is selected such that the angle of deviation is precisely related to their distance to the television monitor 1, the right eye 6 seeing only the right frame 2, the left eye 7 seeing only the left frame 3, the mask 8 obliterating peripheral vision.

The suggested method to house the wedge prism pair 4 and 5 in comfort for the viewer Is that it is mounted In an opaque wall, suitably shaped for comfortable viewing, either seated or standing, with an inner surface covered with a non-reflective material (such that the viewer sees no distracting reflections from the surface of the television monitor 1). On the other side of the wall the television monitor 1 is positioned - 1 6 accurately, its surface the correct distance (related to the deviation angle) to the wedge prisms 4 and 5. In this way the viewer will see only the stereo-paired information 2 and 3 on the television monitor 1.

Figure 2 Illustrates a possible mounting for the pair of wedge prisms 4 and 5 to accommodate the varying distance between viewers' eyes and the ability to slightly rotate the pair. The wedge prisms 4 and 5 are located in mounts 11 which can slide away from and towards each other (indicated by arrows) by a threaded drive rod 14, rotated (indicated by arrows) by a knob 13. The knob 13 also acts to allow rotation (indicated by arrows) of the disc 12 on which the wedge prism mounts are located, allowing the viewer to make the slight alignments which may be necessary for comfortable viewing.

(Experience in tests has shown that viewers vary in their selection of optimal choice of rotation from the horizontal, and, of course, the precise distance between their eyes.) The disc is located on a plate 15 with attachment holes 16 for securing to the suggested "wall" separating the viewer from the television monitor.

The essential features of this nvenbon, in its first embodiment, which is called the "individual embodiment", ( figure 1) for a single human viewer, are best explained by the following example, with certain additional comments: A. Selected existing motion picture or videotape footage is copied by existing telecne process (well known to those versed In the art) onto a digital videotape medium. (The invention Is however, of course, amenable to the use of motion picture film, videotape or other motion picture media shot and recorded specifically for use with this nvenbon.) The higher the picture quality of telecine method and digital videotape format, the higher will be the quality of the copy. Digital Beta is presumed to be the best format as an international broadcast standard at the present time. However the Invention works at the relatively low picture-quality level of, for example, mini DV, and when the present high cost of formats such as HD reduces, this, and others may be more commonly employed. Future high quality formats, yet to be developed, may also be employed. Original sound may or may not be copied at this stage, and a recessive visual timecode signal may also be integrated within each frame to exactly identify it for the viewer who requires it. - 1 7

B. This invention involves selection of appropriate motion picture or videotape footage. Appropriateness refers to the presence of motion parallax within the frame, produced either by camera movement or various sorts of movement of one or more elements within the frame (Categories 1 to 9 above).

C. The digital videotape material is copied, by employing a Serial Digital Interface (SDI), or any system producing no compression (loss of image quality) Into a computer, stored therein in uncompressed form, to form an identified file.

D. Use is then made of existing digital image manipulation and editing software 1 0 programs.

E. The film frames, now in digital form in the computer, are displayed in the form of two screens 2 and 3, side by side, horizontally, on the monitor 1.

For the highest quality display, a CRT monitor is deployed, with the two screens 3 and 2 to the left and right, abutting each other, or slightly separated by black, and ringed by black round the edges of the monitor 1. A wide screen plasma screen may also be suitable.

F. An important element of this invention Is now employed to make possible the three-dimensional viewing of the original footage.

G. Assume, for the purposes of example, that the original footage was produced by a camera "tracking" to the right, although footage of any other category (see: Categories 2 to 9) could be used. Assume that it was a conventional motion picture film camera running at 24 fps. A "start frame" is selected and placed on the left-hand 3 of the two screens on the monitor 1. Then a frame is selected, called the "second start frame", that is likely to be between one to nine frames adjacent to the "start frame", and is placed on the right-hand screen 2 on the monitor 1. The actual displacement, he. the number of frames, between the "start frame" and the "second start frame" will be dependent upon the rate of movement along the original camera track, and that camera's speed (fps), and, on occasion, the character of the movement of objects being filmed, and could be more than nine frames. (The applicant's research with a wide variety of original material finds the average - 18 displacement (time-delay) is from one to five frames, but this average has a wide, and, of course, skewed standard deviation.) The selection of the frame number between the "start frame" and the "second start frame" will also be the result of an aesthetic decision of how successfully the two chosen frames combine to produce the optimal actual or effective three-dimensional composite Image.

H. If the original camera movement was moving to the left, the "start frame" will be placed on the right of the two screens, and the "second start frame" will be placed on the left of the two screens on the monitor.

I. At this stage In the process, the monitor 1 will have two screens 2 and 3, with Images which are slightly different, effectively each being the view of a single lens displaced laterally some 2.5 - 3" (63.5mm to 76. 2mm), this distance approximating that between the left and right human eyes.

J. The monitor is now viewed (by the creative operator) using a binocular pair of wedge prisms 4 and 5, mounted appropriately to produce comfortable human viewing, and located at a distance from the monitor such that it is In focus and that only the whole of each of the two individual screens 2 and 3 is visible to the right 6 and left eye 7 of the operator respectively.

K. The computer program is now set to "play" and the human viewer watches the pair of synchronous moving images 2 and 3. The left eye of the viewer 7 sees the footage from the "start frame" at the same time as the right eye of the viewer 6 sees the footage from the "second start frame".

L. Because the original camera track produced a horizontal displacement such that at some point, likely to be one to five frames, but possibly more, it recorded a frame being the field of view some 2.5 - 3" (63.5mm to 76.2mm) laterally displaced from the original, the newly displayed pair of images on the computer monitor forms a stereoscopic-pair, such that, upon viewing by this invention, the original footage appears in three dimensions.

M. The output medium and display format will be Digital Versatile Disc (DVD) which will hold a digital copy of the three-dimensionally transformed motion picture film or videotape material held In digital form in the computer's hard - 1 9 dsc memory. The DVD will be employed as the portable presentation source for the composite paired stereoscopic Images 2 and 3 on a wide format television monitor screen 1. DVD has the advantage of being an international standard and DVD players are commonly available. This medium also provides a range of easy user controls, e.g. instant re-start and freeze frame etc. DVD is at present the best quality storage medium commonly and internationally available. The results of this invention may however be viewed on lower quality existing formats, e. g. VIES videotape, or on future higher quality formats yet to be made available. In all cases the viewer of the DVD or other medium will look at a wide-screen, or other, television monitor 1 through the binocular wedge prisms 4 and 5. Other viewing systems, employing, for example, two pairs of precisely aligned front-slvered mirrors will not be described. While having the advantage of eliminating chromatic aberration, such systems require sealing to exclude dust and are considerably more expensive in view of the precision required In location of the four reflecting surfaces.

N. A preferred form of location (see Figure 2) for the wedge prisms 4 and 5 is that they be located in a wall and act as two small windows through which the viewer may look. The housing would be such that the wedge prisms 4 and 5 are exactly aligned and with a deviation angle such that the pair of images 2 and 3 on the wide-screen monitor 1 will be centred in the right 6 and left eye 7 respectively. The whole of the monitor 1 would be housed In a light- proof container so that the viewer experiences no unpleasant reflections. The viewer may sit or stand as appropriate to the design of the museum display.

This type of viewing arrangement is compatible with modern museum exhibit practices. Alternatively the viewer can wear the wedge prisms 4 and 5 as spectacles and approach and view the wde-screen television monitor 1, either seated or standing.

Figure 3 shows an example of the "cinema" embodiment. A large projection screen 31 reflects a stereo-pair, the right-hand frame 2 and the lefthand frame 3 projected from a video projector conventionally from behind and above the audience's - 20 heads. The audience sits in seats (A to K). Each member of the audience wears a pair of wedge prism spectacles individually selected for the appropriate deviation angle according tothedistancesa-s,b-stc-s,d-s,e-s,f-sig-s,h-s,l-s,j-s,k-stwherealb,c,d, etf,gth,i,j, k are the seat positions and s the screen plane. The steps A to M above remain the same, but instead of individual viewing of the pair of screens on a wide-screen monitor, the copy recorded on DVD (or on HO videotape or hard disc player, for example) is projected, such that: N1 An audience may be seated facing a wide reflecting screen on which is projected, by digital projection, the recorded footage.

N2 Members of the audience, in each row of carefully positioned seats (A to K), wear a binocular pair of wedge prisms which are selected with appropriate deviation angle related to the distance from the individual row of seats to the screen (a-s to k-s) and which have a field of view such that only the whole of each of the two individual screens is visible to the right and left eye of each audience member respectively. To stabilize the audience's heads, seats with a moulded headrest are likely to be used. The headrests will enable the members of the audience to hold their heads in a stable position such that their right and left eyes are easily centred on the right and left frames of the stereo-pairs on the screen respectively.

If the footage presented has a sound track, this may be heard by the viewer In at least one of three ways: Either: from the loudspeaker/s of the wide-screen monitor in the "individual embodiment" Or: from loudspeakers associated with the cinema screen In the "cinema embodiment" Or: from headphones integrated with the binocular wedge prisms in either embodiment Or: from headphones associated with each seat.

A limtabon of the stereoscopically viewed footage will occur, for example, in footage containing close-up images of speakers and a synchronous live speech soundtrack, because only one of the pair of frames will have a correct synchroncity between picture and sound, and the speaker will lose "lip sync" in the other. It is however possible to produce, in most cases, a satisfactory compromise by locating the time-delay of the sound track such that it is halfway between the correct synchronicity with the left and right images. Because mouth movements in speech produce large changes in a few frames, the time delay (displacement) employed to achieve the three dimensional condition is likely to be only one, two or three frames, in which case "lip sync" will not be a significant problem.

The improved characteristics of this invention are seen to provide an important resource for museum presentations of archive motion picture film and videotape. The resultant footage has all the values of necessary veracity and documentary integrity, while possessing the significant strengths of the third dimension. The method of displaying two screens at once makes the viewer able to attest to the authenticity of the material, unlike most earlier systems. This feature, plus the obvious novelty for general audiences, and also the likely value placed upon this invention in places such as tourist information centres which employ television monitors, give this invention obvious commercial potential. The wedge prisms allowing the viewing of the two frames, improve upon all previous viewing systems, transmitting 100% of the light, not altering the wavelength, imposing no flicker and exposing the authenticity of the original material.

Due to the great expense involved, a further, third, embodiment employing only motion picture film as the medium, will not be described in detail. Those technically familiar with motion picture laboratory procedures will recognize that well known "optical printing" procedures can produce a motion picture negative and projection print containing time-delayed (displaced) paired sub-frames 2 and 3, which, when projected and viewed through a wedge prism pair, will exhibit all the three-dimensionality heretofore described in the earlier embodiments.

Figure 4 represents a conventional X-ray security scanning apparatus. The body of the apparatus 41 houses the X-ray sources, the luggage to be scanned passes through the apparatus on a conveyor 42. Atop the body of the apparatus are two television monitors, side by side, the right 43, the left 44. On these two monitors, respectively are displayed a stereo- pair of X-ray images, the right 2 and the left 3. The viewer (security inspector) would view the stereo-pairs through the wedge prism pair of spectacles, of - 22 appropriate deviation angle such that the television monitors are at a comfortable distance of some few feet.

All airports, seaports and many other public and private locations employ security measures. Travellers' baggage and the contents of transport vehicles are now routinely scanned by X-ray security scanners in an attempt to locate offensive weapons and illicit goods. The most common method of display of the X-ray information is in the form of a pair of television monitors 43 and 44 displaying an image of the item of luggage, or transport vehicle, from two viewpoints at right angles, one horizontal, one vertical 2 and 3. Existing practice employs two points of view in an attempt to improve identification of contained items and discovery of illegal materials. The two television monitors are generally placed abutting each other so that a single operator may scan both screens with ease. The x-ray information Is presented as a coloured or monochrome two-dimensional image. The skill of the operative is exercised in being able to decode the two-dimensional shapes on either screen, and discriminate between safe and unsafe two- dimensional profiles. Much training is necessary to develop this skill. It is the contention of this application that to be able to view this existing Information In three-dimensions will greatly improve the effectiveness of security operatives without the enormous expense of designing and developing actual three-dimensonal x-ray displays. It is believed that small airports and those In developing countries could be advantaged by the relatively small costs involved in employing the result of this embodiment. It is further predicted that this embodiment can greatly increase the crme-deterrent effect of x-ray security scanners and contribute to a safer world.

The "x-ray embodiment" of this Invention provides an improvement on these existing systems. Using the principles earlier outlined, wherein mobon parallax is the source phenomenon for locating pairs of frames which exhibit three-dimensonality, the following describes the method, hardware and software required.

X-ray scanning security equipment is characterized by the items which range from air passengers' hand luggage to whole large container vehicles entering ferry ships. Hand luggage Is placed on a moving conveyer 42 and passed through an x-ray scanning zone 45. An operator monitors the television screen which displays the individual x-ray images 2 and 3. Many modern systems employ two x-ray sources located at right angles which - 23 produce a pair of scans from top to bottom and side to side of the object, these being displayed on a pair of television screens 43 and 44, generally located side by side for ease of viewing. The present application involves a relatively minor set of modifications to existing facilities and practice, as follows: A1) A bmed triggering circuit, related to the speed of the conveyor 42 carrying the baggage, switches two consecutive x-ray photographs (identified as "H" for the horizontal X-ray source, and "V" for the vertical x-ray source), such that an appropriate motion parallax is achieved between the two "H" images, and, with simple switching, the two "V" images, thereby displayed on a pair of television monitors 43 and 44.

B1) The pair of "H" displays, or the pair of "V" displays, is viewed separately by the operator through a pair of wedge prisms such that the right image 2 is viewed by the right eye, and the left image 3 by the left eye.

C1) The stereo pair 2 and 3 appears as a three-dimensional image for the operator and greatly enhances the identification of contained objects.

While the invention has been illustrated in embodiments performing uses in the viewing of mobon pictures, videotape, and television images employing archive or live sources, and x-ray sources in security scanning contexts, it is understood that other applications such as satellite observation, or in underwater environments, forensic examination, military research, and employment of radiation sources other than x-rays, for example are possible. - 24

Claims (17)

1. A stereoscopic display system for moving images, comprising a pair of display screens arranged side by side, means for displaying on the screens respective ones of a pair of moving stereoscopic images, and at least one viewing device positioned in front of said screens and comprising optical means for each of a pair of eyes, constructed and arranged such that each eye of a person looking through the viewing device will see a respective one of said pair of moving images and the person will perceive a moving stereoscopic image.

2. A stereoscopic display device according to Claim 1, wherein the optical means comprises a prism.

3. A stereoscopic display device according to Claim 1, wherein the optical means comprises at least one lens.

4. A stereoscopic display device according to Claim 1, 2 or 3, wherein the display screens are light-emitting displays.

5. A stereoscopic display device according to Claim 1, 2 or 3, wherein the display screens are regions of a single light-emitting display.

6. A stereoscopic display device according to Claim 4 or 5, wherein the or each display is a cathode-ray tube, a liquid crystal display, or a light-emitting diode display.

7. A stereoscopic display device according to Claim 1, 2 or 3, wherein the display screens are light-reflecting screens, and the means for displaying include a pair of projectors.

8. A stereoscopic display device according to any preceding claim, wherein the viewing device is at a fixed location in front of the screens.

9. A stereoscopic display device according to Claim 8, comprising a plurality of viewers positioned at fixed locations in front of said screens, whereby a plurality of people can view the moving images at the same time.

10. A method of creating pseudostereoscopic moving images from an original sequence of 2-dimensional recorded images, comprising creating a sequence of pairs of images in which the first images of the pairs comprise the original sequence of images, and the other images of the pairs comprise images from the original sequence bme displaced relative to the first images. - 25

11. A method according to Claim 10, characterized in that the original sequence of images result from or represent non-linear movement of the recording camera.

12. A method according to Claim 10 or 1 1, wherein the other image in each created pair of images is a copy of an image in the original sequence a fixed number of frames earlier in the sequence than said first image.

13. A method according to Claim 12, wherein said fixed number is from 1 to 9.

14. A method according to Claim 13, wherein said fixed number is from 1 to 5.

15. A method of X-ray examination of an article, comprising recording a first X-ray image of the article using an X-ray imaging apparatus, displacing the article relative to the imaging apparatus and recording a second X-ray image, displaying the pair of images side by side on a display screen or screens, and viewing the images using a viewing device positioned in front of said screen or screens and comprising optical means for each of a pair of eyes, constructed and arranged such that each eye of a person looking through the viewing device will see a respective one of said pair of images and the person will perceive a stereoscopic image.

16. A method according to Claim 15, wherein the optical means comprises a prism.

17. A method of X-ray examination of an article, substantially as described with reference to Figure 4 of the drawings.

Specficabon HOLCUSLIP1 174.GBP2004 10-19 doc

17. A method according to Claim 15 or 16, wherein the or each display screen is a lght-emitting display.

18. A method according to Claim 17, wherein the or each display is a cathode-ray tube, a liquid crystal display, or a lght-emitting diode display.

19. A stereoscopic display system for moving images, substantially as described with reference to, or as shown in, the drawings.

20. A method of creating pseudostereoscopic moving images, substantially as described with reference to Figures 1 and 2 of the drawings.

21. A method of X-ray examination of an article, substantially as described with reference to Figure 4 of the drawings. -r - 2t

Amendments to the claims have been filed as follows:

1. A method of creating from an original sequence of 2-dimensional recorded images forming a moving image record in which motion parallax is present a se- quence of stereo pairs of images each comprising a left and right image which, when viewed separately by the left and right eyes respectively of a viewer provide the viewer with the perception of a moving image comprising at least some 3-dimensional elements, the method comprising using the original sequence of images as the first images of the pairs, and using as the second images of the pairs images from the original sequence ds- placed relative to the first images by from one to five frames.

102. A method according to Claim 1, which further comprises displaying on a I,< pair of display screens arranged side by side respective ones of the sequence of stereo << pairs, and providing at least one viewing device positioned in front of said screens said viewing device comprising optical means for each of a pair of eyes, constructed and ar ranged such that each eye of a person looking through the viewing device will see a re 15 spective one of said pair of moving images and the person will perceive a moving stereo scopic image.

<< 3. A method according to Claim 2, wherein the optical means comprises a . . . prism.

4. A method according to Claim 2, wherein the optical means comprises at least one lens.

5. A method according to Claim 2, 3 or 4, wherein the display screens are light-emitting displays.

6. A method according to Claim 2, 3 or 4, wherein the display screens are regions of a single light-emittng display.

7. A method according to Claim 5 or 6, wherein the or each display is a cathode-ray tube, a liquid crystal display, or a lght-emitting diode display.

8. A method according to Claim 2, 3 or 4, wherein the display screens are light-reflecting screens, and the means for displaying include a pair of projectors.

9. A method according to any of Claims 2 to 8, wherein the viewing device is at a fixed location in front of the screens.

Specification HOLCUSLI P1174 GBP2004-10-19 doc Am

10. A method according to Claim 9, comprising a plurality of viewers pos'tioned at fixed locations in front of said screens, whereby a plurality of people can view the moving images at the same time.

11. A method of X-ray examination of an article, comprising recording a first X-ray image of the article using an X-ray imaging apparatus, displacing the article relative to the imaging apparatus and recording a second X-ray image, displaying the pair of m- ages side by side on a display screen or screens, and viewing the images using a viewing device positioned in front of said screen or screens and comprising optical means for each of a pair of eyes, constructed and arranged such that each eye of a person looking through the viewing device will see a respective one of said pair of images and the person i, will perceive a stereoscopic image.

12. A method according to Claim 11, wherein the optical means comprises a prism. . ,

. ' 13. A method according to Claim 11 or 12, wherein the or each display " 15 screen is a light-emitting display.

14. A method according to Claim 13, wherein the or each display is a cath ode-ray tube, a liquid crystal display, or a lght-emitting diode display.

15. A stereoscopic display system for moving images, substantially as de scribed with reference to, or as shown in, the drawings.

16. A method of creating stereoscopic moving images, substantially as de scribed with reference to Figures 1 and 2 of the drawings.