WO1996024117A1

WO1996024117A1 - Enhancing two-dimensional image to three-dimensional image

Info

Publication number: WO1996024117A1
Application number: PCT/AU1996/000047
Authority: WO
Inventors: Paul Desmond Roche; Robin Mark Yerbury Ford
Original assignee: Trannys Pty. Ltd.
Priority date: 1995-02-01
Filing date: 1996-01-31
Publication date: 1996-08-08
Also published as: AUPN087195A0

Abstract

The present invention relates to a method, apparatus and enhanced image produced by the method or apparatus, whereby two-dimensional images (4, 5, 6) or images composited from at least one two-dimensional image are enhanced so that they can be viewed in conjunction with a lenticular screen to view a three-dimensional effect. The method particularly includes the generation of perspectives from a single two-dimensional image or part of a two-dimensional image, slicing the generated images into segments (1, 1', 1', 2, 2', 2', ...), collating or splicing together each of the segments and then placing the collated image under a lenticular screen.

Description

ENHANCING TWO DIMENSIONAL IMAGE TO THREE DIMENSIONAL IMAGE

Field of the Invention

The present invention relates to image enhancement and more particularly to the enhancement of two dimensional images, so that a 3D effect can be viewed. Background Art

3D photography and the production of 3D images is a well documented art. 3D images are generally produced by taking images of a subject at different angles of view or perspectives. The images can be superimposed and processed so as to be viewed in association with a lenticular screen to give a three dimensional effect to the composited image. The production of 3D images of this sort require the taking of images from many perspectives and processing them to produce the 3D image. This is generally an expensive and complicated exercise because of the difficulties of the setting up of the cameras to take images from different perspectives.

Summary of the invention The invention provides a method of producing an enhanced image or part of an image from an initial 2D image or initial composite 2D image stored in or has been input into a processing means, said enhanced image or part of an image being adapted to be placed under a lenticular screen to view a 3D effect, said method comprising the following steps:

(A) identifying objects in the foreground, centreground or background, or a combination thereof, to be enhanced in a 3D fashion;

(B) generating other perspectives of said foreground, centreground or background objects or a combination of said objects from the initial perspective 2D perspective;

(C) slicing each of the generated perspectives for placement on the final image;

(D) collating or splicing together each of the sliced up portions of the perspectives to form the final image.

It is preferable that the initial image is a photograph, an electronic image, a painting, an artwork, an illustration, a sketch, an architectural drawing, a designer drawing, an animated image, a manufacturing drawing, a plan, an engineering drawing, art illustration, typeface, text or any other 2D representation. It is also preferable that the final image is an electronic representation of an image, or is reproduced onto a video screen, a photo negative a positive transparency, a photographic print, a lithographic separation, or an emulsion attached to a sheet material.

It is preferable that a step between step (b) and (c) is added, that step including the enhancing of objects displayed in said image to compensate for viewing those objects from a different perspective. Further steps can be added to the above method, such steps include : the printing of the image; aligning a lenticular screen to the final image; and bonding the image to a lenticular to produce a final composite 3D enhanced image.

In respect of step (b) or the preferable step between steps (b) and (c) above, the generation or enhancement of the perspectives or objects can include extrapolating that portion which would be located behind a central subject or a foreground subject, because a greater proportion of an object in the image is to be viewed from a different perspective.

Preferably the method also includes providing alignment means around the periphery of the completed enhanced image for cooperation with an aligning step, to check the register of the final image with the lenticular screen.

The invention further provides an enhanced image, which is a combination of 2D and 3D images or just 3D images, said enhanced image being initially a 2D image only with some or all of the objects in the image being processed to produce further perspectives, said further perspectives and said 2D image being spliced and collated to produce a final image, over which a lenticular screen can be placed to view the 3D effect.

It is further preferable that the method can be utilised on type and text in a image whether that type and text has depth, or requires depth to be given thereto. The method can also be used on paintings and art work, illustrations, architectural drawings or plans, animation, manufacturing drawings or plans, designer drawings and any other images that begin or are represented initially as 2D representations.

The invention also provides a method for aligning a collated image with a lenticular screen the collated image having been enhanced and collated from a plurality of images, said method comprising the steps of:

(A) providing an indicator means for each of the plurality of images which go to make up the final collated image;

(B) slicing and transferring a portion of the indicator means together with the sliced portion of each image or portion of an image that is to be collated in the final image;

(C) visually representing the final image and placing a lenticular screen thereon

(D) aligning the lenticular screen with the final image to produce a predetermined enhanced image whereby a desired alignment is indicated by the indicator means when standing at the focal length of the lenticular screen from a predetermined perspective.

Preferably the indicator means is a colour or system of colours associated with each image. Preferably when the image and the screen are in a predetermined alignment the indicator means appears as a single colour. By following the method described above persons responsible for the proper alignment of an image and a lenticular screen have a visual means to check the accuracy of the alignment. If a single colour is associated with each image, then if the border of the final composite image is made up of many colours the alignment is incorrect. The border will have a different colour corresponding to a particular perspective which is being viewed if it is in correct alignment.

This method can be used in association with a colour sensing device which is located at the focal length of the lenticular screen and set up for the appropriate perspective.

The invention also provides for an enhanced image or part enhanced image, said image being formed by means of collating slices of other images and overlying a lenticular screen, said image including a mechanism for checking the alignment of the lenticular screen relative to the collated image.

Preferably the mechanism consists of an indicator means associated with each slice of said other images. It is further preferable that the indicator is comprised of a single colour when the screen is correctly positioned over the collated image and is viewed from a single perspective at the focal length of the screen. Preferably the indicator is a border around the enhanced image. The indicator means can be removed once the screen and image are bonded.

The features of the invention provide a means to convert a two dimensional image into a three dimensional or part three dimensional image by means of extrapolation, and a means to align a lenticular screen with a composited image. Brief description of the drawings

An embodiment of the present invention will now be described by way of example only with reference to the following figures:

Figure 1 illustrates in diagrammatic form, objects in a two dimensional image;

Figure 2 illustrates the objects of figure 1 as generated in a second perspective, being a first stage of enhancement;

Figure 1A illustrates the perspective of figure 1 for the illustration of the method;

Figure 3 illustrates the object of figure 1 as generated in a third perspective, being a second stage of enhancement;

Figures 4, 5 and 6 illustrate the perspectives of figure 1, 2 and 3 in the next stage of enhancement;

Figure 7 illustrates the collated perspectives of figures 4,5 and 6.

Detailed description of the drawings

Illustrated in figure 1 is a rectangular object 10 and circular objects 20 and 30, which are illustrated in 2D as is normal for images and other planar representations. Because it is a 2D illustration moving to either side of the illustration of the objects in figure 1 gives no further information as to the appearance of the respective objects at different perspectives of viewing. For the embodiment of this invention, the initial image can be any two dimensional representation such as a photograph, an electronic image, a painting, a sketch, an architectural drawing, a designer drawing, an engineering drawing, art illustration or any other 2D representation, or an image which is made up of a composite of 2D images

An image of the objects of figure 1 are scanned into a processing means such as a computer. Once the image has been converted into electronic information in a computer, the computer is operated to generate or extrapolate perspectives of the objects at different perspectives to the original object taken in the 2D image of fig 1. This extrapolation or generation of perspectives can be done by mathematical formula or by artistic and/or creative manipulation by the person performing the process as is described below.

Illustrated in figure 2 is a different perspective (namely when a person moves to the left of the objects in figure 1 ) of the objects of figure 1. The perspective of figure 2 is indicated by viewing in the direction of arrow 2 of figure 1. (The illustration of figure 1 is repeated between figures 2 and 3 as figure 1A, to help represent the perspective. The arrows 2 and 3 present in figure 1 have been removed for figure 1A). Paint box or graphics software which allows for the cutting and pasting of pixels or groups of pixels is utilised to produce the perspective of figure 2. An operator uses the pixels which represent the objects of figure 1 and moves them to where the operator believes they should be if viewed from a predetermined desired perspective (extrapolation). If there is any portion missing when pixels are moved the missing portions are simply replaced by similarly coloured pixels as represent the object. Further, shading or other paint box techniques may be required.

Production of a 3D enhanced images can be made using hardware such as the QUANTELL (trade mark) Graphic Paintbox which is a mainframe computer system for production of graphic images. A variety of other software can be utilised to perform this task including

ADOBE PHOTOSHOP (trade mark), SPECULAR COLLAGE (trade mark), ALIAS

SKETCH (trade mark), FREEDOM OF THE PRESS (trade mark) and other graphics and post production software. The selection of the software will depend on the size of the image being manipulated and the amount of other perspectives that it is desired to generate. In the example above only 3 perspectives are utilised to arrive at the destination or final image. In practice anything up to 14 or more perspective images could be utilised. If 14 images were utilised the amount of computer memory necessary to process and produce a final destination image would amount to approximately 2 gigabytes on hard disk and a further 2 gigabytes of auxiliary storage. One initial image, or enhanced image could occupy in excess of 80 megabytes of memory.

The sophisticated software available in today's market can be manipulated to produce these images from different perspectives. A perspective to the right of that of figure 1 can be produced in like fashion to that of figure 2. The example will be continued with reference to only three images namely a left perspective(figure 2), a front-on perspective(fιgure 1) and a right hand side perspective(figure 3), whereas in practice to obtain a smooth transition between perspectives 14 or more is preferred. The left hand side perspective (figure 2), front-on perspective (figure 1) and right hand side perspective (figure 3) are each in existence, electronically in the computer or processing means. At this point either the same software or different software can be selected, but the software must incorporate a cut and paste facility. The software program is manipulated by an operator to firstly slice each picture up into vertical columns as is illustrated in figures 4, 5 and 6. The number of vertical columns will be dependent upon the size of the lens of the lenticular screen that will be used, and the size of the desired final destination image. An example of a lenticular screen is one which has a lens spacing of .62 mm and thus if three images are used each perspective is divided up into .2067 mm columns. The calculations are done on a spreadsheet software so that formulas exist and by feeding in different parameters, the slicing up proportions and spacings can be readily determined. The millimetre spacings can be represented in terms of pixel widths. If 14 perspectives were to be sliced up to fit under a .62 mm lenticular lens, each perspective would be sliced up into columns which are 2 pixels in width.

The first composite column is made up from columns 1', 1 and 1" ( which are from the left hand side perspective (figure 2), front-on perspective (figure 1) and right hand side perspective (figure 3) respectively). The individual columns 1', 1 and 1" are placed side by side in the destination image (figure 7). The second column 2', 2 and 2" of the left perspective (figure 2), front-on perspective (figure 1 ) and right hand side perspective (figure 3) respectively, are also placed side by side, and so on, until the nth column is placed. The first and second composite or sets of columns will take up the first and second lenses respectively, of a lenticular screen. This is done until the whole width of the destination image (figure 7) is composited. Then this is printed out or otherwise visually represented. When one looks at the printed destination image without a lenticular screen, it appears to be an out of focus representation of the objects 10, 20 and 30. At this stage the lenticular screen is added to overlay the composite destination image (figure 7). Because of the optical effect of diffraction of light passing through the lenticular screen and the fact that each eye is seeing two different images standing front on to the composited image, (at the focal distance of the lenticular lenses), a viewer will see a three dimensional image. In this situation a viewer will probably see will see the centre perspective (figure 1) with the left eye and the right perspective with the right eye. When the viewer moves to the left, the viewer will view left hand perspective (figure 2)with the left eye and the centre perspective with the right eye. In this way a 3D image is formed. When the above process is utilised in respect of an object which is placed in the background of a two dimensional image, because it is an image in only two dimensions, further information needs to be generated and stored with the image in the computer because parts of objects that were hidden in a 2D shot will become visible in certain 3D perspectives and visa versa. In the example of figure 1, 2 and 3, the change in perspective from figure 1 to figure 2 will mean that portions of balls 20 & 30 become visible on the left hand side of the ball whilst portions on the right hand side of the balls 20 & 30 move out of view. This will also account for the visual effect that will occur in reality which results in an apparent relative displacement or shifting of the objects due to the change in perspective. Because of the state-of-the-art computer programs and software available this can be painted in when the left hand perspective (figure 2) is generated in this situation.

One of the difficulties with the combination of a composited 3D enhanced image and use thereof with a lenticular screen is the alignment of the lenticular screen with the image.

One method of solving this problem is that when each of the 3 perspectives (or more) of the above example are electronically produced they are stored within a coloured border which is exclusive to that perspective. Thus the perspective of figure 2 has a border 4 which could be coloured red, for example; the perspective of figure 1A (repeated figure) has a green border 5 for example; and figure 3 has a blue border 6 for example. When the perspectives are sliced and then spliced into the destination image (figure 7) the colour specific segment of the border is moved as well. The result is that a destination image having a variety of colours in the border will result. Because the slices from each perspective are put in the same relative positions the colours in the border will have a repeated pattern. When the lenticular screen is placed over the print of the destination image, an operator will be able to tell if it is correctly aligned because when the composited or destination image is viewed from the perspective of figure 1A a green border for example will be viewed at the top and bottom. If there is any other colour in existence or parts of the border have another colour then the lenticular screen is not properly aligned. It may require shifting or rotating, to a very small extent to place the final image and the lenticular screen into correct alignment. This process is readily adaptable to use with a colour sensing device which would be located at a set perspective axis at the focal length of the lenticular lens and calibrated to sense whether a particular colour is present all around the border. If more than one colour is present a signal is sent to the operator and an adjustment made. A border can also be placed vertically at the side of the final destination image.

The processes mentioned above are not at this stage automatic processes. These are performed by operators who are able to cut and paste the original image so as to reposition elements thereof to thereby gain a different perspective. At the moment this is labour intensive, however it is envisaged that by virtue of software development the method can be automated so that simply the parameters of the picture, the parameters of the end dimensions of each of the respective strips, the number of pixels per strip and the number of strips per lenticular lens can be made or calculated and processed to produce the destination or final image.

The above method can be utilised in such a way that, for example, only the foreground and the background of an object are three dimensionally enhanced whereas the object itself is not. This would equate to the balls 20 and 30 of figure 1 being 3D enhanced whilst the rectangular object 10 remains in 2 D. This will give a unique effect of combining 3D with non-3D and if there is any type or text also associated with the image this can also be combined as 2D overlaid or 3D overlaid onto the picture. Because a composite image having 2D and 3D will have a foreground and a background, three lines of text, for example can be respectively located in the background, the object plane and the foreground. The text can also be given depth by the same process as described above, or it can be depicted in 2D at different depth locations in the same destination image.

The above detailed description of the drawings illustrates the method utilising vertical slices and vertically oriented lenticular lenses. Other effects and similar effects can be achieved by the use of horizontally oriented slices and lenticular lenses.

The method can be used to produce a print and thus front lighted or as a transparency and therefore back lighted. If back lighted the image needs to be displayed behind a light box or some other light source in order to achieve a desired effect. Depending on whether front or back lighted a person skilled in the art will need to adjust the parameters and dimensions of the slices to account for these variables and the characteristics of the lenticular screen. The method can also produce colour separations for litho printing, or it can be printed directly on to a lenticular screen which has associated with it an emulsion layer. Alternatively the final image can be printed onto receiving sheets and adhering these to a lenticular screen. Another alternative is to visually produce the final image on a video screen and associate and register a lenticular screen therewith.

Another industrial application of this technology is the use with respect to signage and in particular bilingual and multilingual signage. With the methods above it is possible to replace the different perspectives with representations of words in other languages so that a viewer in one orientation will view see the words of one language whilst at another location the words of a different language will be viewable. When this application is performed the sign is set up so that both eyes will view the same information and in this way there will not be a three dimensional visual effect.

The above describes an embodiment of the present invention and modifications can be made by those skilled in the art without departing from the scope of the present invention.

Claims

1. A method of producing an enhanced image or part of an image from an initial 2D image or initial composite 2D image stored in or has been input into a processing means, said enhanced image or part of an image being adapted to be placed under a lenticular screen to view a 3D effect, said method comprising the following steps:

(B) generating other perspectives of said foreground, centreground or background objects or a combination of said objects from the initial perspective 2D perspective; (C) slicing into image segments, each of the generated perspectives and the initial image or portions of the initial image, for placement on the final image;

(D) collating or splicing together each of the image segments of the perspectives to form the final image.

2. A method as claimed in claim 1 wherein said initial 2D image or said initial composite 2D image is a photograph, an electronic image, a painting, an artwork, an illustration, a sketch, an architectural drawing, a designer drawing, an animated image, a manufacturing drawing, a plan, an engineering drawing, an art illustration, typeface, text or any other 2D representation.

3. A method as claimed in any one of the preceding claims wherein the final image is an electronic representation of an image, or is reproduced onto a video screen, a photo negative a positive transparency, a photographic print, a lithographic separation, or an emulsion attached to a sheet material.

4. A method as claimed in any one of the preceding claims wherein between steps (b) and (c) a further step is added which includes the enhancing of objects displayed in said image to compensate for viewing those objects from a different perspective.

5. A method as claimed in any one of the preceding claims wherein any one of or a combination of the following further steps are added: the printing of the final image; aligning a lenticular screen to the final image; or bonding the final image to a lenticular screen.

6. A method as claimed in any one of the preceding claims wherein there is included an additional step of providing alignment means around the periphery of the completed enhanced image for use in an aligning step, to check the register of the final image with a lenticular screen.

7. A method as claimed in any one of the preceding claims wherein said initial image or said initial composite image includes type and text in an image whether that type and text has depth, or requires depth to be given thereto.

8. An enhanced image produced by the method of any one of claims 1 to 7.

9. An enhanced image, which is a combination of 2D and 3D images or just 3D images, said enhanced image being produced from one initial a 2D image or composite 2D image only, with some or all of the objects in the image being processed to produce further perspectives, said further perspectives and said 2D image being spliced and collated to produce a final image, over which a lenticular screen can be placed to view the 3D effect.

10. A method for aligning a collated image with a lenticular screen, the collated image having been enhanced and collated from a plurality of images, said method comprising the steps of: (A) providing an indicator means for each of the plurality of images which go to make up the final collated image;

(B) slicing and transferring a portion of the indicator means together with the sliced portion of each image or portion of an image that is to be collated in the final image; (C) visually representing the final image and placing a lenticular screen thereon

(D)aligning the lenticular screen with the final image to produce a predetermined enhanced image whereby a desired alignment is indicated by the indicator means when standing at the focal length of the lenticular screen from a predetermined perspective.

11. A method as claimed in claim 10 wherein the enhanced image is adapted to produce a 3D effect with lenticular screen.

12. A method as claimed in claim 10, wherein the enhanced image is adapted to produce a single image via the lenticular screen when viewed from a first perspective and a different image when viewed from a second perspective.

13. A method as claimed in any one of claims 10 to 12 wherein said indicator means is a colour or system of colours associated with each image.

14. A method as claimed in any one of claims 10 to 13, wherein when the image and the screen are in a predetermined alignment the indicator means appears as a single colour.

15. A method as claimed in any one of claims 10 to 14, wherein a colour sensing device, located at the focal length of the lenticular screen and set up for a predetermined appropriate perspective, is used to detect the required colour or the indicator means for the predetermined perspective.

16. An enhanced image or part enhanced image, being formed by means of collating slices of other images and overlying a lenticular screen, said image including a mechanism for checking the alignment of the lenticular screen relative to the collated image.

17. An enhanced image or part enhanced image as claimed in claim 17 wherein the mechanism consists of an indicator means associated with each slice of said other images.

18. An enhanced image or part enhanced image as claimed in claims 17 or 18 wherein the indicator is comprised of a single colour when viewed from a single perspective at the focal length of the screen, when a lenticular screen is correctly positioned over the collated image.

19. An enhanced image or part enhanced image as claimed including one of claims 17, 18 or 19 wherein the indicator is a border around the enhanced image.

20. An enhanced image or part enhanced image as claimed in- claim 17, 18,19 or 20 wherein said indicator means is removed once the screen and image are bonded.

21. An apparatus which produces an enhanced image as claimed in any one claim 8, 9 or 16 to 20, or which uses a method as claimed in any one of claims 1 to 7 or 10 to 15 to produced an enhanced image.