Note: Descriptions are shown in the official language in which they were submitted.
wo gs/oo~o 2 1 6 5 4 3 4 PCT/AU94/00298
IMPROVEMENTS IN THREE DIMENSIONAL IMAGERY
THE PRESENT INVENTION relates to a method and an apparatus
for producing imagery with three visual dimensions.
BACKGROUND TO THE PRESENT INVENTION
Three-dimensional imagery is a term that has been in use for
more than a century: during this time, it has developed a
complex range of meanings.
These connotations vary from a general implication of depth
in imagery to particular types of imagery, some of which
entail opposing ideas.
For example, the term is used to describe depth in imagery
acquired from a single viewpoint, without consideration of
other viewpoints.
In contradistinction, other imagery is described as three-
dimensional, in which no visual depth appears, but where
different viewpoints are displayed.
For the purpose of description and definition, the term
"three-dimensional imagery" as used throughout this
specification, is intended to define imagery that:
"simultaneously contains at least two adjacent angles of
view, acquired from points spaced sufficiently about a
common centre, or continuum of common centres, to display
three apparent dimensions within a coherent visual volume;
without any appearance of more than one image."
Most three-dimensional imagery of this type is accomplished
with viewers, visors and spectacles. They include simple wood
W O 95/00880 2 1 6 5 4 3 4 PCT/AU94/00298
and glass devices of last century, like the "Holmes viewer",
to helmets mounting electronic shutters, synchronised to
film, or video, frame changes.
These systems generally function by blocking views left of
the object centre to the right eye, and views right of the
object centre to the left eye.
The disadvantages of these arrangements include having to
use, or wear, them; to limitations on angles of view; to
incompatibilities with individual visual idiosyncrasies.
Such problems have long been impediments to the general
proliferation of three-dimensional imagery.
After viewers, probably, the next most ubiquitous approach is
the lenticular array, an optical grid, dating to the earliest
history of three-dimensional imagery.
Three-dimensional grid contrivances work on a similar
principle to viewers. To greater or lesser extents, grids
isolate views acquired left of object centres to left eyes;
and isolate views right of object centres to right eyes.
The main difference between grid, compared to viewer systems,
is the position of the view differentiating opticals. Grids
have been generally placed in front of imagery, while viewers
are worn or placed in front of the eyes.
A portrait incorporating a grid concept is reported to have
been produced by the Danish painter, S.A. Bois-Clair, in
1692. The painting is described as presenting a row of
narrow, vertical, alternating strips of two views of a
person, each strip separated by a vertical lath.
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
The common concept embodied here is that the wood dividers
mask left side picture strips from right eyes, and right side
picture strips from the left eyes. This segmentation and
separation of two off-set paintings, if actually constructed
well enough to work, would have expressed three-dimensional
grid theory in action. Further, it would have demonstrated
the inherent limitations of simple grid systems.
The first drawback of the design above is the presence-of the
grid itself. It must be prominent enough to block each eye to
half the imagery and is, therefore, just as visible as the
imagery. For this reason the quality of the effect is reduced
to the extent that the grid must be in focus.
A second disadvantage of simple three-dimensional grid
systems is the lack of optical uniformity. For full three-
dimensional imagery to be seen, every element of each view
must be equally evident to the corresponding eyes. This
restricts image sizes and viewing angles to being small,
around the image centres.
In practice, large, high quality, three-dimensional images,
that numbers of observers can view from wide angles, are not
possible for simple static grid arrangements.
Lenticular arrays allow a substantial improvement that
overcomes the first grid defect to some extent; although,
again, generally, for small images. The improvement is
realised by replacing solid grid segments with transparent,
thin, vertical, lens strips. The lenses are angled, so that
left views are in focus to left eyes when right views are in
focus to right eyes.
While most of the glass grid is unseen, because it is
transparent, the edges where the lens strips join are not.
WO 95/00880 2 1 6 5 4 3 4 PCT/AU94/00298
These edges and joins partially obscure the image, as well,
they can introduce other undesirable effects, including
spectral aberrations.
Moreover, all the other limitations of simple three-
dimensional grid systems still apply.
A further partial improvement is available from a dynamic
grid system called "The Cyclostereoscope".
Here, the presence of a grid can be removed from view
completely, by revolving it around a screen at sufficient
speed to make it invisible to the eyes. Such an improvement
was proposed in French patent specification No 607,961 of
~rancoise Savoye, dated October 8, 1942.
This design provided an effective solution to the problem of
grid visibility. Like the lenticular array, the
cyclostereoscope introduced new detractions and did not solve
any more of those characteristics of the simple grid concept.
Again, the improvement was confined to small images,
presenting partial three-dimensional effects.
The mechanical limitations of the cyclostereoscope have been
overcome by improvements described in international patent
specification No PCT/AU92/00199 and Australian patent
specification No PL6295.
These improvements provide three-dimensional imagery that can
be viewed from a wide angle without wearing visors, or
similar personal view differentiating opticals.
wo 95,00880 2 1 6 5 4 3 4 PCT/AU94/00298
SUMMARY OF INVENTION
According to one aspect of the present invention there is
provided a grid arrangement for use in three-dimensional
imagery, including a grid formation formed of a plurality of
spaced apart slats between a viewer and said imagery, and
wherein said slats are laterally spaced apart one from the
other, having dimensions of both width and length.
According to a further aspect of the present invention there
is provided a grid arrangement for use in three-dimensional
imagery, including a grid formation formed of a plurality of
spaced apart slats between a viewer and said imagery, and
wherein said slats are laterally spaced apart one from the
other, having dimensions of both width and depth; said grid
arrangement being formed so that the spacinqs between said
slats are tapered.
It should be appreciated that in all forms of the invention,
the grid arrangement of the present invention can be separate
from or incorporated into an appropriate screen arrangement,
or alternatively can be programmed into an appropriate
computer software package or programme to provide the grid
for achieving the objects of the present invention.
Alternatively, any appropriate known or available mechanical,
electrical or liquid crystal means may be used to form the
inventive grid arrangement of the present invention.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
The invention will be described by way of example only and
with reference to the accompanying drawings, wherein:
Figure 1 is a view which demonstrates a simple,
static, optical grid arrangement,
W095/04880 2 1 6 5 4 3 4 PCT/AU94/00298
Figure 2(I) illustrates a similar arrangement to
Figure 1 on a different scale,
Figure 2(II) illustrates a similar arrangement to
Figure 2(I) but distinguished by a
larger viewing distance,
Figure 2(III) illustrates a similar arrangement to
Figure 2(I) but distinguished by a far
greater viewing angle,
Figure 2(IV) illustrates a similar arrangement to
Figure 2(I) also distinguished by a
greater screen width,
Figure 3 demonstrates an arrangement where
there is a divergence of conditions in
Figures 1 and 2,
Figure 4 demonstrates a further arrangement
with a greater width of image segments,
Figure 5 demonstrates an arrangement with small
image segments separated equally,
Figure 6 illustrates an arrangement
representing two images of adjacent
angles of view,
Figure 7 illustrates an arrangement showing a
short fall in a simple static grid
arrangement,
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
Figure 8 illustrates an arrangement
demonstrating two sections of grid
elements containing vertical elements,
Figure 9 demonstrates an arrangement of grid
sections with vertical depth,
Figure 10 demonstrates another arrangement of
the present invention,
Figure 11 demonstrates a further arrangement of
the present invention,
Figure 12 illustrates a further arrangement of
the present invention and in
particular a situation common to
optical grid systems, and
Figure 13 illustrates an arrangement according
to a further form of the present
invention.
DESCRIPTION OF THE INVENTION
This invention provides a number of improvements to three-
dimensional imagery that can be produced by transposing
optical grids before imagery containing discernibly displaced
adjacent angles of view.
Preferably, the adjacent angles of views are displayed in
separated segments, alternatively, and transpose fast enough
for the transposition to be unnoticeable for the human eye.
Preferably, the transposition of the separated, alternate,
image segments is substantially synchronous with transposing
wo 95,00880 2 1 6 5 4 3 4 PCT/AU94/00298
grid elements. The grid segments are preferably shaped,
profiled and positioned, to reveal from any viewing position
substantially every element and aspect of the imagery.
Adjacent angles of views left of object centres are revealed
only to the left eye of an observer, while substantially
simultaneously, all imagery acquired right of object centres
is revealed only to the right eye of an observer.
The invention permits three-dimensional imagery to be seen
without visors, from a wide angle, without limiting the image
area viewed.
The invention further permits three-dimensional imagery to be
seen without visors, from any distance without limiting the
image area viewed.
The present invention allows for any desired viewing
position, every element and aspect of imagery, acquired left
of object centres, to be continuously and completely revealed
to the left eyes of observers, while simultaneously,
continuously and completely, obscuring every element and
aspect of- that left of object centre imagery from their right
eyes. Substantially simultaneously, every element and aspect
of imagery acquired right of object centres is revealed to
the right eyes of observers. Again, simultaneously, as well
as again continuously and completely, every element and
aspect of this right of object centre imagery is obscured
from their left eyes.
The present invention further preferably provides that the
minimum angle required to produce visibly different images
can be varied. This variation is without limitation, to the
maximum displacement possible at the object acquisition
distance, for optimum three-dimensional effect. Additionally,
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
the variation is provided without restricting the areas of
imagery available to either eye.
The present invention further provides for the display of
vertical three-dimensional effects, simultaneously with
horizontal three-dimensional effects, without restriction.
The present invention further provides focal planes within
the imagery to be in focus simultaneously. If required
otherwise, focal planes can be in combinations of focal
conditions, without limitation.
The present invention further provides an improvement which
permits the apparently seamless joining of individual images,
whether three-dimensional, or two-dimensional. This
improvement permits imagery of non-standard format (such as
on wide, large, curved, or special purpose screens), to
further enhance realism, or produce other special effects.
The shape and positioning of grid segments are important to
the success of the present invention and will now be
described and discussed further as follows:
Known grid technology
It is well known that grid patterns can be placed in front of
imagery containing adjacent angles of view to produce three-
dimensional effects.
Until now, these effects have been limited and restricted
from application with advantage to television, computer
screens, video monitors and many other forms of imagery in
common use.
W O 95/00880 2 1 6 5 4 3 4 PCT/AU94/00298
Improvements that have been incorporated in three-dimensional
imagery grid systems include lens array adaptations, usually
in lenticular forms, and dynamic grid arrangements.
Neither of these two approaches satisfy requirements for
general imagery systems applications.
In case of lens, and lenticular, arrays, a major drawback is
focal point sensitivity which limits viewing to a small arc
around the centre. Practical application requires
inordinately large, and expensive, screens to give an
adequate viewing angle to even small audiences.
Inertial forces and other fundamental incompatibilities of
mechanical grid systems with conventional electronic screens
are overwhelming disadvantage for systems like the
cyclostereoscope.
At least some of these impediments are overcome or at least
minimised by improvements set out in international patent
specification No PCT/AU92/00199 and in Australian patent
specification No PL6295.
This provides three-dimensional imagery that can be viewed
from a wide arc and at large distances in relation to image
size.
As well, these improvements lend themselves to take advantage
of the physiognomy of normal human perception.
For instance, the propensity of the eyes to focus on imagery,
tends to enhance its perceived quality, whatever the defects
caused by grids in front of it.
wo gs/oo~o 2 1 6 5 4 3 4 PCT/AU94/00298
The ability of the mind to retain partial visual impressions,
additively for cumulative recognition, helps perception of
grid systems imagery, if it is partial, or sequential.
As well, normal eye movement can contribute substantially to
the horizontal scanning essential to all such grid designs.
The improvements set out in international patent
specification No PCT/AU92/00199 and Australian patent
specification PL6295, together with the natural tolerance and
compensation of human vision, provide an extremely flexible
system that can appear to work very satisfactorily, well
beyond technical optimums.
These optimums determine the quality of three dimensional
imagery. As well, they define the limits of image sizes;
viewing angles; viewing distances; applications; audience
sizes; compatibility with human vision, and conventional
equipment.
For the purposes of this invention, these technical optimums
apply when -
i): All imagery left of object centres is at the maximumangle of divergence possible from all imagery
acquired right of object centres for any object
acquisition distance, and
ii): All imagery acquired left of object centres is seen
by left eyes only, and completely; while all imagery
acquired right of object centres is seen by right
eyes only, and completely.
It will be appreciated that these criteria are incompatible
with simple static grids. Therefore, an improved dynamic grid
W O 95ioo880 2 1 6 5 4 3 4 PCT/AU94/00298
arrangement is provided which is essential to acquire the
desired standard(s) in addition to the other features of the
present invention, as referred to above.
Simple static grid arrangement
In the accompanying drawings, referred to herein, the
following symbols are used:
L Left eye position
R Right eye position
al a2 a3 a4 a5 a6 Image segments seen by left eye
bl b2 b3 b4 b5 b6 Image segments seen by right eye
c c c c c Grid segments
d Image width
e Viewing distance
f Grid distance from the screen
Figure 1 demonstrates a simple, static, optical grid
arrangement.
A left eye position, L, is spaced from a right eye position,
R, at typical pupil separation of two and a half inches.
The eye positions, L and R, face screen AB, fifteen inches
wide; at a viewing distance, e, of thirty inches. Grid
segments at c, cl, c2, C3, C4 and C5 separate the viewing
arcs from positions L and R to image segments al, bl, a2, b2,
a3~ b3~ a4~ b4~ as, bs~ a6 and b6.
The view from left eye position, R, is confined completely to
image segments al, a2, a3, a4, a5 and 6
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
The view from right eye position, L, is confined completely
to image segments bl, b2, b3, b4, b5 a 6
The grid segments, separate, and isolate views from position L
exclusively to position L; and segments, separate, and
isolate views from position R exclusively to position R.
The width and position of the grid segments permit
unrestricted viewing from position L, of image segments al,
a2, a3, a4, as and a6-
Similarly, the width and position of the grid segments permitunrestricted viewing from position R, of image segments bl,
b2, b3, b4~ bs and b6-
This arrangement demonstrates imagery segmented and separated
in alternate, vertical, strips for complete and separate
viewing from two positions, without restriction on the total
area of any image segment.
For this simple, static, optical arrangement, grid sections
in focus will be seen.
Grid sections of less width than those at c, cl, c2, C3, C4
and C5 can be positioned at greater distances from imagery on
the screen AB, with identical effects to those grids segments
at positions c to C5.
An image of the width above - 15 n - with a viewing distance
of 30", could be typical of a video monitor for data or word
processing, however, the screen width can be any width, and
the viewing distance, any distance at which imagery can be
discerned.
It can be seen, therefore, that within the constraints of
normal vision, a screen of any width can be positioned at a
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
distance where, grid segments of correct width, can be
located to completely and exclusively separate left and right
eye views of alternate segments of imagery of equal width and
area.
Similarly, a screen can be positioned, at any distance, or
angle from where imagery can be discerned, and grid segments,
of correct width, located to completely and exclusively
separate left and right eye views of alternate segments of
imagery of equal width and area.
Figure 2 demonstrates that these conditions apply for any
simple, static, grid arrangement of this type.
Figure 2 I illustrates a similar arrangement to Figure 1 on a
different scale. Here, eye positions L and R face image
segments al, bl, a2, b2, on screen AB; twenty-five inches
wide; at a viewing distance e, of five feet. This situation
could be typical of household television viewing.
Figure 2 II illustrates a similar arrangement to figure 2 I,
but distinguished by a far larger viewing distance. Here, eye
positions L and R face image segments a3, b3, a4, and b4 on
screen CD; twenty-five inches wide; at a viewing distance,
el, of ten feet.
Figure 2 III illustrates a similar arrangement to Figure 2 I
again, but distinguished by a far greater viewing angle.
Here, eye positions L and R, face image segments a5, b5, a6,
b6, on screen EF, twenty-five inches wide, at an angle of
45o.
Figures 2 IV illustrates in similar arrangement to Figure 2
I, also; distinguished by a greater screen width. Here, eye
wo gs/oo~o 2 1 6 5 4 3 4 PCTIAU94/00298
positions L and R, face image segments a7, b7, a8, b8, on
screen GH, sixty inches wide.
Such an arrangement could apply to a high definition
television screen.
As can be seen in all these examples, grid sections can be
placed at a position where alternate image segments are
completely and equally separated to the corresponding-eyes,
and each eye sees its corresponding image segments completely
and equally.
It can be seen therefore, that these conditions apply to all
image sizes, viewing distances, and angles of view.
Figure 3, demonstrates a divergence from the conditions in
Figures 1 and 2, where grid sections positioned at c and cl,
only partially separate views from eye positions a and R of
image segments a, b, al, and bl. In contrast, grid segments
positioned at c2, and c3, completely separate the views from
eye position L and R exclusively.
Figure 4 demonstrates that the greater the width of the image
segments to be separated to the left and right eyes, the
further from the screen the position for the grid segments
for complete, equal, separated, and exclusive viewing by the
corresponding eyes.
Figure 5 demonstrates that very small image segments can be
separated equally, and completely; as well as viewed
completely, separately, and exclusively, by corresponding
eyes; when grid sections are close, or very close to image
segments.
W095/00~0 PCT/AU94/00298
21 65434
16
Here, eye positions L and R view image segments al, bl, and
a2, 0.6 inches wide; at a viewing distance, el, of 9 feet. It
can be seen that the grid sections c, cl, and c2 can be
positioned close, very close, or almost coincident with image
segments al, bl, and a2 on screen AB.
Eye positions L and R at viewing distance el, of 4.5 feet,
and angle to the screen AB of 45, require essentially
identical grid segment positions for separate and exclusive
left and right eye views of corresponding image segments al,
bl and a2 to those for left and right eye positions L and R.
Eye positions L and R of viewing distance e2, twelve feet,
from image segments a, bl and a2, on screen AB, require grid
segment positions to separate exclusive left and right eye
views of the image segments, again, essentially identical to
those for left and right eye positions L and R.
It can be seen, therefore, that where alternate image
segments are small, small grid segments can be placed very
close in front of the image segments, to provide complete,
completely separated, and exclusive views to left and right
eyes of separated, alternate, image segments, over large
viewing distances, and at wide angles of view.
Further, the smaller the image segments, the smaller the grid
segments required, and the flexibility of viewing positions.
It can be seen that the arrangement depicted in Figure 5
could apply to any similar arrangements, including any
viewing distances, or angles of view where image segments can
be seen. Consequently, it can be seen, too, that for small
and very small image segments, there are no limitations on
the viewing positions through corresponding small grid
segments, where, within the bounds of human vision
discernment, complete, completely separate and exclusive, are
w09sioo880 2 1 6 5 4 3 4 PCT/AU94/00298
views provided for the left and right eyes of corresponding
alternate segments of imagery, and where each eye sees
exactly half of the total image segments.
It follows that it is possible to position grid segments in
front of any uniform imagery, that can be seen, so that the
eyes see, see completely, and exclusively, equal numbers of
image segments.
It also follows that it is possible to position small grid
segments close to imagery, so that the eyes see, see
completely and exclusively, equal numbers of image segments
from wide angles and various viewing distances.
It is essential for all these arrangements that the width and
area of image segments are the same; and the width and area
of the grid segments are the same.
This requirement is incompatible with the application of
simple, static, grids to three-dimensional imagery.
The displacement of corresponding visual elements, in images
containing adjacent angles of view about a common centre,
increases from the image centre.
Figure 6 represents two images, of adjacent angles of view
about a common centre, of equal height and width, displayed
on a screen. IL is imagery acquired left of the object
centre; IR is imagery acquired right of the object centre.
In any such arrangement, the geometric vertical image centre
lines C, Cl, C2, and C3 can be aligned coincidentally.
With vertical centre line C and Cl, aligned coincidentally with
C2 and C3, all image elements offset from the geometric
vertical image centres will diverge according to the angle of
W095/00~0 2 1 6 5 4 3 4 PCT/AU94/00298
divergence between the two images, the distance of the image
element from the centre line and in the direction of the
image centre from the common object centre.
In Figure 6, e is an image element of image IL, an image
acquired left of the object centre.
The distance, d, between image element, e, and image vertical
centre C, Cl, will be greater than the distance, dl, between
image element, el, and the image vertical centre C2, C3.
Similarly, the distance d3, will be greater than d2.
As well, both the distances, d2, and d3, will together be
greater than the distances d, and dl.
It can be seen, therefore, that corresponding image elements
of imagery containing adjacent angles of view about a common
centre, cannot be aligned to coincide at more than one
position,; and the discrepancy in coincidence of non-aligned
points increases from the image centre; as well as with
increasing divergence between the image centres.
From this, it can be seen that it is not possible to place
grids of equal segment width in front of alternate segments of
imagery containing two adjacent angles of about a common
centre; so that the grid segments separate the view acquired
left of the object centre to the left eye; and the view
acquired right of the object centre to the right eye, in each
case exclusively, and in each case so that all image segments
are seen completely by the corresponding eye.
Obviously, it is possible in the case of a simple, static,
grid to place grid segments of unequal length, so that image
segments, containing imagery acquired left of object centres,
WO95/00880 2 1 6 5 4 3 4 PCT/AU94/00298
are seen completely and exclusively by the left eyes; and
image segments, containing imagery acquired right of the
object centres, are seen completely and exclusively by the
right eyes.
However, the arrangement above is visible from only one
viewing position; so it is not possible for the arrangement
to provide the required exclusive and complete separation
from any position.
Figure 7 shows another failing of the simple, static, grid
when applied to imagery containing discernibly different
adjacent angles of view.
If grid segments are of equal width, positioning in front of
imagery containing adjacent angles of view, acquired at wide
divergences from the common centre, can result in each eye
seeing both left and right views simultaneously.
In Figure 7, grid segment c, placed between image segments a
and b2, completely and exclusively separates the left and
right eye positions L and R to the corresponding image
segments al and a2, for L and bl for R.
Image element el appearing in image segment al and seen by
the left eye position L, does not have a corresponding image
element in image segment bl, to be seen by the right eye.
Image element el and e2 appear in image segment al, because
of the greater distance from the image centre line in
proportion to the distance of e2 from the centre of line C.
This is a typical example of so-called "double imaging" in
partial three-dimensional imagery arrangements.
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
The following improvements overcome these limitations.
The following individual improvements to known technology can
be applied in combinations of each individual improvement
according to the final optical and three-dimensional quality
required.
The first of these is to apply vertical dimension to dynamic
grid segments.
Dynamic grid segments with vertical dimension
Transposing grid sections in front of alternate segments of
imagery, containing visibly distinct adjacent angles of view,
averages the blocking effect of each grid segment along its
path. If the speed of transposition is sufficient to be
unseen by the eye, the grid segments disappear and produce a
view of the imagery which is a total of the average blocking
effect of each grid segment along its path of travel.
Positioning the grid close, or very close, to the image
segments limits the extent to which the eyes can see around
the grids, producing the effect of vertical depth, or
dimension, in the grids, and to that extent a "tunnel view".
This improves the effectiveness with which the grid elements
accurately separate the left and right image segments to the
corresponding eyes.
Where scope to improve left-right separation by reducing the
distance of the grid from the screen is constrained, then,
the efficiency of the grid can be enhanced by increasing the
physical depth of the grids.
Increasing the physical dimensional of each grid segment, by
the vertical component extending away from the eye and in the
wo 95,00880 2 ~ ~ ~434 PCT/AU94/00298
direction of the imagery, creates a "tunnel view"- effect
between any two grid segments, that can completely separate
views of each image segment to the corresponding eye,
providing the vertical extensions are long enough.
Figure 8 of the drawings demonstrates two sections of grid
elements Gl and G2, both containing vertical elements
extending from the viewer forward the screen, and where Gl is
closer to the screen than G2.
As can be seen, vertically extended grid sections G2 separate
both left and right views from L and R completely and
exclusively, while vertically extended Gl do not.
The depth and width of the grid sections can be sized,
according to the width of the image segments, and the
required position of the grid from the screen.
If the grid sections are of equal depth and width, and the
same width and height as the image sections, then grid
sections containing vertical depth can be placed any distance
from the image segments to completely segment and separate
the image segments to corresponding eyes, providing the grid
segments have both appropriate depth and width.
Grid sections of appropriate depth and width for image
segments, containing discernibly different adjacent angles of
view, can be placed to separate images completely to the
respective eyes to produce three-dimensional imagery.
Transposing the grid elements, in synchronisation with
transposing image segments, will produce a three-dimensional
image, in which the grid segments cannot be see, providing
the speed of transposition is sufficiently fast.
W095/00880 2 1 6 5 4 3 4 22 PCT/AU94/00298
Increasing the angle of view in three-dimensional imagery
formed by dynamic grid sections with vertical depth
Increasing the vertical depth of dynamic grid sections for
the production of three-dimensional imagery, decreases the
viewing angle at which the imagery can be seen unless the
grid sections containing vertical depth are shaped to
maximise the angle of view.
Figure 9 demonstrates an arrangement of grid sections with
vertical depth, having a tapered form to permit wide angle
viewing. Such tapered shapes can be either equilateral or
isosceles, and of size depending on the angle of view
required, the size of the image and image segments, and the
viewing distances involved.
It may be preferred to have the grid shapes easily adjustable
for different situations.
As well as having tapered edges facing observers, the grid
sections with vertical depth may taper in the other
direction, for instance if the screen is curved. As well, the
grid segments can be oval, instead of wedge shaped, or
diamond shaped, according to viewing requirements.
Transposing grid elements containing vertical depth so as to
reveal all image segments completely to the appropriate eyes
Figure 10 demonstrates that for any one position of the grid,
parts of the image segments may not be seen; resulting in a
partial view of the imagery. This applies particularly when
grid segments contain vertical depth.
A complete view of each image segments can be obtained by
either of two methods.
wo gs/oo~o 2 1 6 5 4 3 4 PCT/AU94/00298
Firstly, by transposing the grid segments along a continuous
path of length equivalent to at least the horizontal width of
each grid segment, all grid segments being of equal width.
The transposition may be continuous, in one direction, or
oscillating.
Secondly, by transposing the grid segments as changes between
fixed positions, such as positions formed on an electro-
optical display, like a liquid crystal display panel or~
similar inertia free means of forming a dynamic optical grid.
In the second case, where grid sections transpose between
fixed points, it is important that the distance of
transposition is equivalent to the width of the grid
sections, and that each individual movement of the grid
segments, within the total transposition, is no larger than
the length of the widest piece of imagery obscured from view
by the grid, at any position, from the required viewpoint of
maximum angle from the image centre.
Where the number of steps that the grid segments can take in
any completion of the total path of transposition, is limited
by available refresh rate, then further alternatives are to
vary the grid positions at a speed at which they cannot be
see, or vary grid segment widths, in a fixed ascending, or
descending ratio, according to the number of movements in the
complete cycle, as shown in Figure 11.
Where the shape, or position, of the grid is changed during
the full transposition cycle, the alternating segments of
imagery behind the grid must change in exactly the same way
simultaneously.
W095/00880 21 6 5 4 3 4 PCT/AU94/00298
24
Dynamic forty-five degree segments containing vertical depth
Figure 12 depicts a situation, common to optical grid
systems, where a fixed viewpoint, encompassing the positions
L and R, can, because it is fixed, result in only partial, or
sequential views of imagery being seen at that point.
Where the transposition of grid segments passes a fixed point
during the transposition cycle, such as the position of an
eye, or a position between eyes, then the view from that
position will be constant. This can result in a left, or,
right view only being seen from that position, or left and
right view sequentially, or no view of an area of imagery.
A solution to this problem is provided by a grid arrangement,
where grid segments angled at 45 to the horizontal,
transpose before alternate segments of imagery, sized shaped,
and angled identically to the grid segments, and transposing
also, in synchronisation with the grid segments, at a speed
where the transposition is invisible to the eyes, to produce
simultaneous vertical and horizontal scanning or oscillation
of the imagery. This arrangement, including vertical scanning
of imagery, also permits the production of three dimensional
imagery containing both horizontal and vertical separation.
Figure 13 illustrates this arrangement, demonstrating that
any two eye views, from any position, will simultaneously
contain both left views for the left eye, and right views for
the right eye, and in no circumstances can either eye be
restricted to partial, or sequential views.
In all circumstances, these provisions can be provided by a
grid system placed in front of imagery containing discernibly
different adjacent angles of view, where the grid transposes
at a speed sufficient to be invisible to the eyes.
WO95/00880 2 1 6 5 4 3 4 PCT/AU94/00298
Alternatively the grid system can be incorporated within a
layered screen, where the image segment perform the function
of grids and imagery, alternately, or simultaneously.
Together, these arrangements provide the display of three-
dimensional imagery of maximum angle of divergence for any
acquisition distance through a combination of features of the
present invention, which include:
1. transposing grid segments, before synchronously
transposing alternate segments of imagery, containing
distinctly different adjacent angles of view about common
centres, all at a speed to render the transposition invisible
to the eye: where -
2. the grid segments should be positioned close or veryclose to the image segments, or have vertical depth,
sufficient to separate the two adjacent angles of view
exclusively and completely to corresponding eyes; and where -
3. the grid segments are tapered to provide maximumangle of view in any direction; and where -
-
4. grid and image segments change in shape, or position
during transposition, so that no grid or image segment
repeats any position in any complete cycle; and where -
5. the grid segments are angled at forty-five degrees to
give simultaneously, both vertical and horizontal scanning
or alternation of the imagery, also appearing in forty-five
degree alternate segments, and where -
6. grid and image segments always maintain equalcorresponding areas on the screen and identical corresponding
shape at any instant of any transposition.
W O 95/00880 21 6 54 34 26 PCT/AU94/00298
The provision of unrestricted focus within three-dimensional
imagery
Real objects have no focal points; normal vision involves
human eyes focusing at will through continuously changing
positions.
This provision can be provided in three-dimensional imagery,
produced as described above by the following means:-
By varying the focal distances, within the total field of theimagery, from the objects closest to the acquisition point to
the objects farthest from the acquisition point. The focal
points may vary continuously with frame changes, or have a
set variation, or change in any desired manner.
Where the imagery is generated artificially and not recorded
from real life, focal planes can be set, or can change
continuously according to the result desired.
It should be appreciated that at all times the adjacent
angles of view contained within the imagery should remain of
identical size and that objects within the imagery remain the
same size even if in different focal positions.
In practice, this effect can be produced by automatically
varying the foci of the camera lenses with frame changes so
that the lenses scan back and forth through the field of view
continuously during recording.
Seamless joining of images
The joining of different images on a screen has well known
difficulties that are caused primarily by the uneven
W095/00880 2 1 6 5 4 3 4 PCT/AU94/00298
illumination of image edges that are always very visible and
in practice impossible to hide from view.
In an arrangement involving a grid system appearing before
the imagery it is a simple matter to ensure that image edges
are always positioned to lie behind grid positions where the
edges would be invisible.
It should be appreciated that in the present invention any
form of mechanical and/or electrical and/or electronic means
can be used or applied to bring into effect the present
invention. In particular, the grid arrangement of the
present invention can be brought about by any mechanical,
electrical, electronic or other means. For example,
mechanical means, electrical means, liquid crystal screen
means, or computer generated means such as a computer
programme generated to provide the grid system within the
screen of a viewing arrangement (such as for example a screen
or television screen), but still so as to provide the good
arrangement between the imagery as viewed and the viewer.
It should be appreciated that improvements and modifications
may be made to the invention without departing from the
scope thereof as defined by the appended claims.
