JP2008244835A - Device and method for displaying three-dimensional image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a three-dimensional display device freely changing over a two-dimensional image display mode and a three-dimensional image display mode and conducting an operation, by easily using a proper display method. <P>SOLUTION: The three-dimensional display device has a memory buffer for writing an input image, a capturing and acquiring the image written to the image buffer, an image converting section image-converting a format in a manner that the image obtained by the capture is converted into a 3D image from a 2D image and an overlay buffer writing the image-converted image. The three-dimensional display device, further has a display displaying the image written to a main buffer or the overlay buffer, and a changeover section changing over the display of the image written to the main buffer and the overlay buffer by the 2D image or the 3D image in the input image. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an apparatus and a method for displaying a stereoscopic image.

In recent years, various types of stereoscopic displays have been developed and put into practical use. It has been possible for several decades to enjoy two-dimensional photographs by optically observing the two images on the left and right with the left and right eyes, and the beam reproduction of the original object like a hologram. The technology to display stereoscopic images by faithfully reproducing the image is also old. Recently, in combination with an electronic display, an environment where various stereoscopic image contents such as moving images can be enjoyed is being created.

For example, there is a stereoscopic image display device using a vertical lenticular lens. This includes 18 subpixels (that is, 6 pixels) in one pitch of the lens, and light rays from the 18 subpixels are distributed in 18 directions of the observation space by the lens. By displaying the light beams thus separated from all the lenses, a stereoscopic image display device in which different images are observed depending on the position of the eyes becomes possible. This method is called an integral imaging method, and has a feature that a natural stereoscopic image can be displayed. (For example, see Patent Document 1)
JP 2006-98779 A

The image output to the stereoscopic image display device needs to be in a unique format. Since images viewed from multiple viewpoints need to be interleaved and displayed, for example, a certain pixel of an image from a certain viewpoint is arranged as three subpixels arranged vertically at a certain place on a stereoscopic image display device Is done. The interleaved image cannot be observed as a correct image with the lenticular lens removed. In other words, this indicates that a normal two-dimensional image cannot be observed as a normal image when observed through a lenticular lens.

A stereoscopic image display system is often constructed on some kind of computer system such as a PC (personal computer). In a PC, a system using a window (hereinafter referred to as a window system) is used as a typical OS, but these are, of course, composed of two-dimensional images. Therefore, the window displayed when the window system is operated cannot be correctly observed, and a serious problem remains in the operability.

The present invention has been made in consideration of the above-described problems, and is a stereoscopic display device or a stereoscopic display that easily switches between a two-dimensional image display mode and a three-dimensional image display mode and performs an operation using an appropriate display method. An object is to provide a display method.

According to the present invention, a main buffer for writing an input image, a capture for acquiring the image written in the image buffer, and an image conversion for converting the format of the image acquired by the capture from a 2D image to a 3D image The main buffer, a display for displaying the image written in the overlay buffer, whether the input image is a 2D image or a 3D image, and the main buffer. There is provided a stereoscopic image display device comprising a switching unit for switching display of an image written in the overlay buffer.

By these means, it is possible to easily switch between the two-dimensional image display mode and the three-dimensional image display mode and easily perform work using an appropriate display method.

Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS.
(First embodiment)
FIG. 1 shows a configuration of a stereoscopic image display apparatus 100 according to the first embodiment of the present invention. The stereoscopic image display apparatus 100 includes a switching unit 101, an image buffer 102, a display unit 103, a capture 104, and an image conversion unit 105.

Here, a system using a general window (hereinafter referred to as a window system)
In this system, a PC is operated, and a 3D image (3D image) is displayed in the environment. The window system itself is a two-dimensional image (2D image), on which a stereoscopic image display application is launched to display a stereoscopic image. Further, when the stereoscopic image display application is not used, it is necessary that the window system itself, menus, icons, and the like can be observed as a two-dimensional image.

First, an image buffer 102 for storing the displayed image data is associated with the window system, and a two-dimensional image constituting the window system is written into the main buffer 1021 of the image buffer 102. The capture 104 acquires a two-dimensional image of the window screen from the main buffer 1021. In order to display the two-dimensional image on the three-dimensional image display device, the image conversion unit 105 (hereinafter referred to as 2D-3D conversion) converts the format of the two-dimensional image acquired by the capture 104 and converts the three-dimensional image. An image that can be correctly observed on the display device 100 is created. The created image is displayed on the display unit 103. This creation method will be described in detail later.

At this time, the image converted by the image conversion unit 105 is written into the overlay buffer 1022 different from the main buffer 1021 that has acquired the input image. For example, in a typical window system, in order to effectively display agents (character animations that help use the application) and shadows such as icons and mouse cursors, transparency information is included in the window system. A so-called layered window is prepared to be displayed over the associated main buffer 1021. In order to overlay and display on the main buffer 1021 by this layered window,
The 2D-3D converted image is written in the overlay buffer 1022, and the display unit 103
Is displayed. At this time, if the transparency is set to 0 (the image written in the main buffer 1021 is not transmitted), the original window screen cannot be seen, and the stereoscopic image display device 1
Only the image converted so that it can be observed at 00 is displayed.

As described above, there are several merits in acquiring an original image from the main buffer 1021 and displaying the converted image on the layered window. First, this 2D-3
D conversion can be implemented as a part of an application program, but there are few means for capturing a window system screen from an application. However, since there is means for capturing the displayed window, the image from the overlay buffer 1022 is not captured at this time, and the image from the main buffer 1021 can be captured.

Therefore, even if a 2D-3D converted image is displayed in a layered window, 2D-
The 3D conversion means can acquire an image of the next new window system. If 2
If the D-3D converted image is written back on the original main buffer 1021, when the next window system image is acquired, the 2D-3D converted image is acquired. Will be displayed.

Due to the characteristics of the stereoscopic image display device, the effective resolution when stereoscopic display is lower than the number of pixels of the panel. For example, in a stereoscopic image display device of an integral imaging method with vertical 12 parallax, when the number of pixels of the panel is 1920 × 1200 pixels, if a method of assigning parallax in units of sub-pixels is used, the resolution of the stereoscopic image is 480 × 400. 2D-3
Only when displaying after D conversion, a resolution of 480 × 400 or double 960 × 400 is possible. In any case, the original window size is 1920 × 1200 pixels, but the entire screen cannot be displayed simultaneously. So capture a part of it,
The device of displaying on the stereoscopic image display apparatus 100 is required. If you change the capture position by moving the mouse, you can observe the full range of the window system, but not at the same time.

However, it is not appropriate to always enable display of the 2D-3D converted image. When displaying stereoscopic image content, an image that is originally observed correctly as a stereoscopic image through a lenticular lens is displayed, so if it is converted 2D-3D, it cannot be correctly observed as a stereoscopic image. Accordingly, in such a case, it is necessary to invalidate the display of the 2D-3D converted image. Switching of the 2D-3D converted image display is performed by the switching unit 101.

This switching may be explicitly instructed by the user or may be automated. For example,
When an application for displaying a stereoscopic image is started up, the switching unit 101 performs switching so as to invalidate the display of the 2D-3D converted image. When the application ends, the switching unit 101 switches to enable display of the 2D-3D converted image. Further, instead of starting / ending the application, the playback start / stop of the stereoscopic content may be triggered.

This is because the stereoscopic image display application also has a number of menus and the like, and it may be more convenient to perform 2D-3D conversion display when operating them. By including such a switching unit 101, it is possible to switch the mode automatically / manually at an appropriate timing, and to display appropriately whether it is a two-dimensional image or a three-dimensional image content. .

Here, details of the 2D-3D conversion will be described. In the stereoscopic image display apparatus 100,
The 3D image content has a format in which images corresponding to the number of parallaxes are combined on a tile, and an image (elements) in which element images that are parallax-decomposed by one lenticular lens are arrayed by appropriately converting the images. By obtaining an image array) and observing it through a lenticular lens, stereoscopic viewing can be achieved. This will be described.

FIG. 2 is a perspective view schematically showing the entire stereoscopic image display apparatus 100. The display device that displays a stereoscopic image shown in FIG. 2 has a flat display unit 33 that displays a parallax composite image as a flat image.
1 is provided. A lenticular sheet 334 shown in FIG. 3A is provided on the front surface of the flat display unit 331 as a parallax barrier 332 for controlling light rays from the display unit 331 or FIG.
A slit plate 333 shown in FIG. Here, lenticular sheet 334
Alternatively, the slit plate 333 is collectively referred to as a parallax barrier 332. Here, the parallax barrier includes an optical aperture. If the parallax barrier is a lenticular sheet 334, the optical aperture corresponds to each cylindrical lens. If the parallax barrier is a slit plate 333, the optical aperture is This corresponds to a slit provided in the slit plate 333. The optical aperture of the parallax barrier 332 substantially restricts light rays from the display unit 331 directed to the viewing area where the stereoscopic image is displayed, and forms a two-dimensional image displayed on the display unit 331. It is provided corresponding to each element image. Therefore, the parallax composite image displayed on the display unit 331 is composed of a number of element images corresponding to the number of optical apertures of the parallax barrier 332. As a result, the three-dimensional image is displayed on the front surface or the rear surface of the three-dimensional image display device by projecting the element images toward the space in the viewing zone through the optical aperture of the parallax barrier 332.

In the stereoscopic image display device 100, the diffusion sheet 301 may be provided between the flat image display unit 331 and the parallax barrier 332 as necessary. In addition, the parallax barrier 322 is
The flat image display unit 331 may be installed on the back side.

The stereoscopic image display apparatus 100 is a one-dimensional integral imaging method (hereinafter, II method). In this one-dimensional II method, a horizontal parallax 341 is given when viewed from a viewpoint 343 on an assumed viewing distance L. However, a stereoscopic image without the vertical parallax 342 is observed. 4A shows the front surface of the stereoscopic image display apparatus 100, and FIG. 4B shows the arrangement of the optical system in the horizontal plane of the stereoscopic image display apparatus, the element image average width Pe, and the second horizontal pitch. (Horizontal pitch of opening of parallax barrier) Ps, viewing distance L, plotting line (straight line group 346) showing the relationship between viewing zone width W is shown, FIG. 4C is a stereoscopic image shown in FIG. Display unit 3 of the display device
3 schematically shows an angle of view in a vertical plane in a viewing zone space with 31 as a reference.

As shown in FIGS. 2 and 4B, the stereoscopic image display device includes the flat display unit 331 that displays a flat image of a liquid crystal display element and the parallax barrier 332 having an optical opening as described above. Yes. The parallax barrier 332 is a lenticular sheet 334 having a shape in which optical apertures extend linearly in the vertical direction and periodically arranged in the horizontal direction as shown in FIGS. 3 (a) and 3 (b).
Alternatively, the slit plate 333 is configured. In the projection type display device, the parallax barrier 3
32 is constituted by a curved mirror array or the like. In this stereoscopic image display device, the parallax barrier 332 from the position of the eye within the range of the viewing angle 341 in the horizontal direction and the viewing angle 342 in the vertical direction.
Through this, the display device 331 is observed, and a stereoscopic image can be observed on the front surface and the back surface of the display unit 331. Here, the number of pixels of the planar image display unit 331 is the minimum unit pixel that becomes a square.

As an example of counting in groups, the horizontal direction (horizontal direction) is 1920, and the vertical direction (vertical direction)
Is 1200. Here, it is assumed that each minimum unit pixel group includes red (R), green (G), and blue (B) pixels. In this specification, “pixel” means a minimum unit whose luminance can be controlled independently within one frame of the display surface. Red (R), green (G), blue in a normal direct-view transmission type liquid crystal panel. The sub-pixel of (B) corresponds to “pixel”.

In FIG. 4B, a distance (assumed viewing distance) L from the parallax barrier 332 to the viewing distance plane 343, a parallax barrier pitch (horizontal pitch of the optical aperture of the parallax barrier 332) Ps, and a parallax barrier gap d are determined. If so, the width of each element image is determined. That is, the average pitch Pe of the element images is an aperture (optical opening of the parallax barrier 332) from the viewpoint on the viewing distance plane 343.
The center of the aperture is determined by the interval between points projected on the display surface of the display device along a straight line toward the center. Reference numeral 346 indicates a line connecting the viewpoint position and the center of each aperture, and the viewing zone width W is determined based on the condition that the element images do not overlap each other on the display surface of the display device.
As already described, an elemental image is a set of pixels that generate a light bundle that passes through an optical aperture with a parallax barrier 332 and is directed to the viewing zone between the parallax barrier 332 and the viewing distance plane 343. This corresponds to the displayed two-dimensional composite image (part of the parallax composite image). A plurality of element images are displayed on the display unit 331 and projected to display a stereoscopic image.

In the parallel light one-dimensional II system in which the horizontal pitch Ps of the apertures is determined to be an integral multiple of the pixel pitch Pp, the average pitch Pe of the element images contributing to the display of the stereoscopic image determined corresponding to each aperture is the pixel It is not an integral multiple of the pitch Pp, and a fraction is accompanied by this integral multiple. Even in the one-dimensional II system in a broad sense where the horizontal pitch Ps of the aperture is not determined to be an integral multiple of the pixel pitch Pp (which does not form a parallel light beam group), the average pitch Pe of the element image is generally the same as the pixel With fractions deviating from integer multiples of the pitch Pp. On the other hand, in the multi-view method, the average pitch Pe of the element images is determined to be an integral multiple of the pixel pitch Pp. 1D I
In the I method, an integer obtained by dividing the horizontal pitch Ps of the aperture by the pixel pitch Pp is referred to as “number of parallaxes”.

Each element image is composed of a set of pixel columns extracted from the parallax component image 426 corresponding to the direction of each parallel ray group, as will be described with reference to FIGS. 5 (a), (b) and (c). Is done. As is apparent, the parallax composite image for displaying a single stereoscopic image is also a set of element images (also referred to as an element image array), and a set of a large number of parallax component images 426 constituting this element image ( It is also a set synthesized in an interleaved manner.

FIGS. 5A, 5B, and 5C show a method of constructing a parallax composite image based on a parallax component image in the parallel light one-dimensional II system. As shown in FIG. 5A, the displayed object (subject) 421 is projected onto a projection plane 422 that is actually disposed on the plane on which the parallax barrier 332 of the stereoscopic image display device is placed. In one-dimensional IP, the vertical direction is perspective projection,
The projection is performed toward a projection line 425 which is a plane parallel to the projection plane 422 and is set at the center of the plane having a viewing distance L and is directed to the projection center line 423 so that the horizontal direction is parallel projection. In this projection, the projection lines do not intersect each other in the horizontal direction, but intersect at the projection center line in the vertical direction. By this projection method, an object image 424 as shown in FIG. 5B is created on the projection plane 422, with the vertical direction being perspective-projected and the horizontal direction being parallel-projected. The object image 424 shown in FIG. 5B corresponds to an image projected in the projection direction 428 shown by reference numeral 1 in FIG. 5A, and in the one-dimensional II, FIG. As shown in a), a subject image 424 projected in a plurality of directions is required.

As shown in FIG. 5B, a projection image corresponding to an image for one direction in which the vertical direction is perspective-projected on the projection plane 422 and the horizontal direction is parallel-projected, that is, the parallax component image 426 is along the vertical direction. The pixel image is divided into pixel columns, distributed to each element image corresponding to each optical aperture (aperture), and arranged in the parallax composite image 427. The parallax component image 426 is displayed on the display surface 4 of the display device.
27, the distance between the aperture pitches Ps (optical aperture pitches Ps) (
The same number of sub-pixel columns as the number of parallaxes) are spaced apart from each other and are arranged separately from each other.

Each projection direction shown in FIG. 5A corresponds to the parallax direction for observing the parallax component image 426 specified by the parallax number, and each direction is not determined to be equiangular, and the viewing distance plane. Above, the projection centers (camera positions) are set at equal intervals. That is, by projecting the camera by moving the camera on the projection center line 423 at regular intervals (the orientation is constant), the intervals between the projection centers are set at regular intervals.

FIG. 6 is a perspective view schematically showing a configuration of a part of the stereoscopic image display apparatus. A case is shown in which a lenticular sheet 334 made of a cylindrical lens having an optical aperture extending in the vertical direction is arranged as a parallax barrier 332 on the front surface of a flat parallax image display unit such as a liquid crystal panel. The optical apertures of the parallax barrier 332 are not limited to being linearly extended as shown in FIG. 6, but may be arranged and formed obliquely or stepwise. As shown in FIG. 6, pixels 34 having an aspect ratio of 3: 1 are arranged in a matrix in a straight line in the horizontal direction and the vertical direction on the display surface of the display device, and each pixel 34 has the same row and column. The red (R), green (G), and blue (B) are arranged alternately in the horizontal direction. This color arrangement is generally called a mosaic arrangement.

In the display screen shown in FIG. 6, the pixels 34 in 18 columns and 6 rows have one effective pixel 43 (this one effective pixel 43 is indicated by a black frame in FIG. 6), or the pixels 34 in 18 columns and 3 rows. One effective pixel is constructed. With such a structure of the display unit, stereoscopic image display that provides 18 parallaxes in the horizontal direction is possible.

Originally, in order to observe a two-dimensional image with the stereoscopic image display apparatus 100, it is only necessary to create an element image array on the assumption that all parallax images are the same image. As a result, an image as if the plane placed on the display panel is stereoscopically viewed is observed. That is, a plane in a three-dimensional space is observed. At this time, the two-dimensional image is expressed with the same resolution as the stereoscopic image.
As described above, the resolution of the stereoscopic image is lower than that of the original display panel.
For example, in a stereoscopic image display device using a 12-parallax oblique lenticular lens, when the panel resolution is 1920 × 1200, the resolution of the stereoscopic image is 480 × 400. However, this lens is designed as a lens with crosstalk to remove moire,
By using this crosstalk, the resolution 960 is doubled only when a two-dimensional image is displayed.
It is possible to display with × 800. The resolution of 480 × 400 is quite narrow for operating the window system. However, if the resolution is 960 × 800, the window system can be sufficiently operated. When a two-dimensional image is displayed at double resolution, the plane is not a display panel surface, but is expressed as a slightly floating surface or a sinking surface. A detailed method for generating a high-resolution two-dimensional plane will be described.

A configuration of the stereoscopic image display apparatus 100 is shown in FIGS. FIG. 7A is a plan view showing the configuration of the stereoscopic image display apparatus. The stereoscopic image display device 100 is a two-dimensional image display device 1.
And a light beam control element 2. The two-dimensional image display device 1 is a liquid crystal display device, for example, and includes a display surface that has a plurality of pixels and displays a two-dimensional image. The light beam control element 2 is provided on the front surface of the two-dimensional image display device 1 and has a plurality of lenses, and controls the directions of light beams from the plurality of pixels on the display surface.

In the stereoscopic image display device, when displaying a character display or a two-dimensional image, the distance zopt from the light beam control element 2 to the display position of the two-dimensional image satisfies the following conditions.

0 <zopt <L / (1 + D) / 2 (1)
Here, L indicates the distance from the observer to the light beam control element 2, that is, the viewing distance, and D is called a depth factor, and is obtained from the following equation (5).
D = L / 8 / (L / 2 / lp) 2 / tan (θ) / pp (2)
Here, lp represents the lens pitch of the light beam control element 2, pp represents the pixel pitch, and θ represents half the viewing zone angle. Hereinafter, derivation of the formula (1) and the formula (2) will be described.
As described above, the II system uses a binocular parallax to distribute stereoscopic light by distributing light rays to a plurality of predetermined parallax angles by a light beam control element regardless of the position of an observer from a certain opening. It is a method to make it. When using a lenticular lens as a light control element,
As a method for assigning parallax images in FIG. 7A, the lens 2 is placed by placing the lenticular lens 2 on the front surface of the display device 1 and placing the focal point of the lens at the pixel position of the display device 1.
With this curvature, a parallel light beam corresponding to the angle is emitted from one lens 2. At this time, assuming that the viewing zone angle is 2θ and the number of parallaxes is N, the viewing zone angle per parallax is

Viewing zone angle per parallax = 2θ / N (3)
It becomes.
By the way, it was considered that only one ray from adjacent parallax usually enters the pupil. 1
Entering more than this number is called crosstalk and causes deterioration in image quality. However, the liquid crystal display has been improved in definition, and multi-parallax of 10 parallax or more can be realized, and the viewing zone angle per parallax has been reduced. In addition to the increase in the number of parallaxes, the light beam spreads in the human pupil 4 in consideration of the pixel width of the liquid crystal display device as the two-dimensional image display device 1, the diffusion film for preventing moire, and the defocus of the lens. An adjacent parallax image can also be seen.

In FIG. 7A, reference numeral 11 indicates a parallax image on the display surface of the two-dimensional image display device 1, and reference numeral 19 indicates a two-dimensional character or image displayed in the pop-up area. In image 11,
There is a parallax image composed of RNA, MA, LNA, RNB, MB, and LNB. MA is a main image of a lens 2, RNA is an image adjacent to the right side of the main image MA, and LNA
Is an image adjacent to the left side of the main image MA. MB is a main image of a lens adjacent to the lens 2 having the image MA as a main image, RNB is an image adjacent to the right side of the main image MB, and LNB is an image adjacent to the left side of the main image MB. .

FIG. 7B shows the ratio of the image of the principal ray 7 and the image 4 of the adjacent parallax entering the pupil 4.
FIG. 7B shows a case where two adjacent parallax image rays that are in contact with the principal ray are visible with the principal ray having the originally assigned parallax angle as the center as the ray from one lens. Let x (parallax) be the number of parallaxes including the principal rays seen between adjacent principal rays, Y0 (= 1 parallax) is occupied by the principal ray 7, and Y1 (parallax) is occupied by one adjacent parallax image. ,

Image by principal ray: Image by adjacent parallax ray = 1: 2 * Y1 = 1: (x−1)
It becomes. As the amount of crosstalk, the ratio of only the adjacent parallax images
Crosstalk amount = x-1 / x (4)
It becomes. As can be seen from FIG. 7B, when information on a point for interpolating adjacent principal ray points is displayed at a position where only adjacent parallax images can be seen, characters with improved resolution can be seen.

First, when the adjacent parallax image interpolates the principal ray 7, the adjacent parallax ray must pass at least on the side closer to the principal ray than the midpoint of the adjacent principal rays 7 in FIG. This is because if the deviation t between the adjacent parallax ray and the principal ray at the stereoscopic display position is larger than half of the lens pitch lp, the adjacent parallax ray 9 from the adjacent lens intersects at the position of the eye of the observer 4. This is because the image looks like a double image. 7A, the optimal position zopt for displaying a two-dimensional character or image is the position at which the adjacent parallax ray 9 from a certain lens intersects with the principal ray 7 from the adjacent lens zn. Then,

zopt <zn / 2 (5)
The condition to become is necessary. Here, how to obtain zn will be described. Therefore, as a definition of zn, when a three-dimensional display is viewed at a viewing distance, the resolution determined from the lens pitch (Nyquist frequency) is the same as the resolution determined from the light density emitted from one lens. Since the parallax angle between the principal ray 7 and the ray 9 of the adjacent parallax image is always the same, a two-dimensional character or image can be obtained not only from the front but also by changing the viewing angle.

Depth factor D is D = αimax / βnyq (6)
It is defined as zn satisfies the following expression in the case of the pop-out amount.
αimax × zno / | L-zno | = βnyq (7)
Here, the viewing distance L is shown in FIG. 7A, and the viewing distance is the distance between the observer 4 and the light beam control element 2. The viewing distance is displayed on a two-dimensional display device in order to display a three-dimensional body image.
The calculated visual distance when creating the dimension mapping data is a value unique to the image data and does not indicate the position of an arbitrary observer. Here, in the pop-up area,

βnyq ＝ βimax
The position zn becomes as follows by changing the expression (7) from the condition of L-zno> 0:
D × zno / (L-zno) = 1
That is,
D × zno = L-zno
Therefore,
zno = L / (1 + D) (8)
It becomes. When the reference is rewritten not to the position from the observer 4 but to the projection amount zn on the display surface or from the light beam control element 2 below the display surface.
zn = L · zno = L × D / (1 + D)
The pop-out amount that satisfies the condition is substantially equal to zn shown in FIG.
At the maximum pixel pitch
D = L / 8 / (L / 2 / lp) 2 / tan (θ) / pp (9)
It becomes. That is, equation (3) is obtained.
Therefore, by substituting equation (8) into equation (5), the optimal position of the two-dimensional character is as described above.
0 <zopt <L / (1 + D) / 2
D = L / 8 / (L / 2 / lp) 2 / tan (θ) / pp
It becomes.
Until now, it has been described that the pop-up position is determined so that the adjacent parallax image is a position for interpolating between the principal rays in the pop-out area.
Next, a description will be given with reference to FIG. 8 that the three-dimensional characters displayed in the pop-out area using the specific characters and the image entering the eyes of the observer are in the same direction on the left and right, so that the display is smooth and easy to see. To do. The depth limit of the character display area will be described later.

In FIG. 8A, when two-dimensional characters or two-dimensional images are displayed in the pop-up area, where they are projected on the parallax number on the two-dimensional display device 1, and on the observer's pupil 4 I showed how it looks. However, the parallax number of the principal ray 7 is 6 parallaxes, the adjacent left parallax number is 5 parallaxes, and the adjacent right adjacent parallax number is 7 parallaxes. The center lens is B, the left lens is A, and the right lens is C.

On the liquid crystal display which is the two-dimensional display device 1,
5A 6A 7A ... 5B 6B 7B ... 5C 6C 7C
7A 6A 5A ... 7B 6B 5B ... 7C 6C 5C in the 3D display object in the pop-up area.
Corresponds to the position of. Since the positional relationship is maintained even at the position of the eyes of the observer, the light rays that enter the pupil 4 include the adjacent parallax image data. 7A 6A 5A ... 7B 6B 5B ... 7C 6C 5C
Thus, a correct interpolation image can be seen on the stereoscopic display object. In FIG. 8A, reference numeral 3 denotes a region where adjacent parallax rays are within half of the interval between the principal rays, and reference numeral 5 denotes the principal ray 7.
Indicates an area that is mainly visible, reference numeral 6 indicates an area in which adjacent parallax is mainly visible, reference numeral 9 indicates a locus of the central part of the adjacent parallax light beam, and reference numeral 10 indicates the main light beam 7 due to the spread of the adjacent parallax light beam. 11 indicates a region, and 11 indicates a parallax image displayed on the display surface of the two-dimensional display device 1.
3 indicates a position where a two-dimensional character is displayed and its parallax number assignment, and reference numeral 15 indicates a position where the adjacent parallax image and the principal ray 7 intersect.

In FIGS. 8B, 8C, and 8D, for example, when the letter E is represented, 2
It also shows how it is displayed on the dimension display device 1 (see FIG. 8B) and how it appears in the eyes of the observer (see FIG. 8D). As shown in FIG. 8C, the chief ray passes through only one of the letters E, but if the adjacent parallax rays are included, three parallax rays pass through the letter E. FIG. 8 (12) shows a pattern image obtained by projecting them onto the two-dimensional display device. In one lens of the lenticular lenses, the left and right parallax images are reversed. When these are viewed from the observer's position through the lens, they can be observed as normal characters as in the characters shown in the pop-out area as shown in FIG. Further, it can be seen that the adjacent parallax image can be recognized as a character by supplementing the principal ray.

Here, the processing flow of the first embodiment will be described with reference to FIG. For example, when the image input by the switching unit 101 is switched in response to activation / termination of an application or 3D content reproduction start / stop (S901), in the case of a 2D image, the capture 104 is the main of the image buffer 102. The 2D image written in the buffer 1021 is captured (acquired) (S902). In order to display a two-dimensional image on the stereoscopic image display device, the image conversion unit 105 converts the 2D image captured by the capture 104 into a 2D-3D image (S903), and can be correctly observed on the stereoscopic image display device 100. Create an image like this. The created image is displayed on the display 103 so that it can be observed on the stereoscopic image display device 100 (S904). On the other hand, in the case of a 3D image, it is not necessary to perform 2D-3D conversion and is displayed on the display 103 as it is (S904).

As described above, the 3D image display application also has a number of menus and the like. When operating these, it may be more convenient to display 2D-3D converted images. Therefore, the mode can be switched automatically / manually at an appropriate timing, and it is possible to appropriately display a 2D image or a 3D image content.

(Modified Example)
Although the configuration examples described so far are relatively easy to implement, the performance may not be so high. This is because data acquisition from the window system image buffer may not be performed at such a high speed. There are several configurations to achieve better performance:

As a form of the embodiment, the switching unit 100 like the stereoscopic image display device 100 shown in FIG.
1, a capture 1002, and a display 1005, and a display 1
005 is a method of incorporating a 2D-3D conversion function including the display driver 1003 and the display unit 1004 into the display driver.

When the switching unit 1001 switches to a 2D image, the 2D image acquired by the capture 102 is written in the 2D image image buffer that configures the display driver 1003a. The image conversion unit 1003b converts the format of the 2D image to generate a 3D image in order to display the 2D image written in the image buffer for 2D image on the stereoscopic image display device. The generated 3D image is written into the 3D image image buffer 1003c and displayed on the display unit 1004.

On the other hand, the 3D image is written in the 3D image image buffer 1003c and displayed on the display unit 1004 in the field blue that is switched to the 3D image by the switching unit 1001. For this purpose, the display driver itself is configured exclusively for the stereoscopic image display apparatus 100.

As another embodiment, the stereoscopic image display apparatus 100 shown in FIG. 11 includes a switching unit 1001, a capture 1002, and a display 1005. The display 1005 includes a display driver 1003, a display unit 1004, and a filter. This is a method of incorporating a 2D-3D conversion function including the driver 1006 into the filter driver.
Depending on the type of OS, the driver can be configured in a layer structure, and a driver called a filter driver can be inserted before the display driver. Then, 2D-3D conversion is implemented at the filter driver level.

When the switching unit 1001 switches to a 2D image, the 2D image acquired by the capture 102 is written in the 2D image image buffer that forms the filter driver 1006a. The image conversion unit 1006b stores the 2 written in the image buffer for 2D images.
In order to display the D image on the stereoscopic image display device, the format of the 2D image is converted to generate a 3D image. The generated 3D image is written in the 3D image image buffer 1006c and displayed on the display unit 1004 via the display driver 1003.

On the other hand, the 3D image is written in the 3D image image buffer 1003c and displayed on the display unit 1004 in the field blue that is switched to the 3D image by the switching unit 1001. For this purpose, the display driver itself is configured exclusively for the stereoscopic image display apparatus 100.

In the above two embodiments, there are restrictions on the functions that can be used inside the driver.
When processing with heavy calculation load such as 3D conversion is performed, performance may not be achieved.
In such a case, it is also possible to create a 2D-3D conversion part as a separate program, send data to the program from the filter driver, and perform 2D-3D conversion.

As another example of the embodiment, the stereoscopic image display apparatus 100 shown in FIG. 12 includes a switching unit 1001, a capture 1002, and a display 1005. The display 1005 further includes a display driver 1003 and a display unit 1004. There is a method using a virtual driver realized by a virtual driver 1007 and an image conversion program 1008. A virtual driver is a virtual driver that does not have actual hardware. The window screen displayed on the virtual display driver is transferred to the image conversion program, and the two-image conversion program draws the converted image on the display.

With such a configuration, it is possible to easily switch between the two-dimensional image display mode and the three-dimensional image display mode and easily perform an operation using an appropriate display method.
(Second embodiment)
Next, a second embodiment will be described. In the first embodiment, 2D-3D conversion is performed to 2
Either a three-dimensional image is output to a stereoscopic display device, or a stereoscopic image content is output as a stereoscopic image. However, as a whole, a two-dimensional window system is properly displayed on the stereoscopic image display device, and when the stereoscopic image display application is operating in the window mode, the stereoscopic image is also correctly displayed. In some cases, the usage is desirable. This is achieved by the method described below.

A display driver 100 like the stereoscopic image display device 100 shown in FIG.
3, a display unit 1004, a filter driver 1006, and further a filter driver 1
006 is a 2D image buffer 1006a and an image conversion unit 1006b for 2D images.
The 3D image image buffer 1006c includes a coordinate conversion unit 1006d for 3D images.

The 2D image is written in the image buffer for 2D image constituting the filter driver 1006a. The image conversion unit 1006b receives the 2D written in the image buffer for 2D images.
In order to display the image on the stereoscopic image display device, the 2D image is format-converted to generate a 3D image. The generated 3D image is written in the 3D image image buffer 1006c and displayed on the display unit 1004 via the display driver 1003. Thus, it is displayed as a 2D image on the stereoscopic image display device.

On the other hand, the position information of the 3D image display portion of the 3D image display application is input to the filter driver 1006, and the 3D image display portion does not perform 2D-3D conversion. In the 3D image display application, a 3D image that has undergone format conversion for displaying a 3D image is directly input to the display driver without being subjected to 2D-3D conversion processing by the filter driver 1006. At this time, the coordinates of the window position on the desktop at the pixel of the 3D image display and the display position on the display screen are different. Therefore, the position information input to the filter driver 1006 is converted into position information on the display screen by the coordinate conversion unit 1006d and displayed.

Here, as a display example, as shown in FIG. 13B, for example, a stereoscopic image display device using a lenticular lens for oblique 12 parallax will be described. Panel resolution is 1920 x 120
When 0, the stereoscopic image resolution is 480 × 400. The resolution that can be displayed by 2D-3D conversion is 960 × 800. Now, the desktop size of the window system is 960 × 800. 2D-3D conversion is implemented using the filter driver 1006, and the desktop itself is 2D-3D converted and displayed as a two-dimensional image on the stereoscopic image display device. Position information of the 3D image display part of the 3D image display application is input to the coordinate conversion unit 1006d of the filter driver 1006, and the 3D image display part is 2D-.
3D conversion is not performed. After the 3D image display application performs format conversion for displaying a 3D image, the filter driver 1006 does not perform 2D-3D conversion processing and is input to the display driver 1003 as it is. At this time, the coordinates of the window position on the original 960 × 800 pixel desktop and the display position on the display screen are different. Therefore, the coordinate conversion is performed by the filter driver and displayed at an appropriate position.

With such a configuration, it is possible to easily create a content combining 2D images and 3D images without creating content for 3D images more than necessary. In addition, since an existing window system based on a 2D image can be appropriately displayed and a stereoscopic image can be displayed on the window system, a 3D image display system can be easily constructed based on an existing PC system. .

The block diagram showing a 1st Example. The perspective view which shows the whole schematic of a three-dimensional image display. The perspective view which shows a lenticular sheet and a slit board roughly as a parallax barrier. The figure which showed the three-dimensional display image typically. Explanatory drawing which shows the structure method of the parallax composite image based on the parallax component image in a parallel light 1-dimensional II system. The perspective view which shows roughly the structure of a part of stereo image display apparatus. The figure which shows the structure of a three-dimensional image display apparatus. The figure explaining where it was projected on the parallax number on a two-dimensional display device, when a two-dimensional character or a two-dimensional image is displayed. The figure which shows the processing flow of a 1st Example. The block diagram showing another Example. The block diagram showing another Example. Configuration diagram showing another embodiment Configuration diagram representing the second embodiment

Explanation of symbols

100 stereoscopic image display device 101 switching unit 102 image buffer 103 display 104 capture 105 image conversion unit 1021 main buffer 1022 overlay buffer

Claims (5)

A device for displaying a stereoscopic image,
A main buffer to write the input image,
Capture to capture the image written to the image buffer;
An image converter that converts the format of the image acquired by the capture from a 2D image to a 3D image;
An overlay buffer for writing the image converted image;
A display for displaying an image written in the main buffer or the overlay buffer;
A stereoscopic image display apparatus comprising: a switching unit that switches display of an image written in the main buffer and the overlay buffer depending on whether an input image is a 2D image or a 3D image. A device for displaying a stereoscopic image,
A switching unit that switches depending on whether the input image is a 2D image or a 3D image;
Capture for acquiring a 2D image switched by the switching unit;
An image buffer for 2D images for writing an image obtained by the capture;
An image conversion unit that converts a format from a 2D image to a 3D image;
An image buffer for 3D images to which the image converted by the image conversion unit or the 3D image switched by the switching unit is written;
A stereoscopic image display apparatus comprising: a display unit that displays an image written in the image buffer for 3D images. A device for displaying a stereoscopic image,
A switching unit that switches depending on whether the input image is a 2D image or a 3D image;
Capture for acquiring a 2D image switched by the switching unit;
An image buffer for 2D images for writing an image obtained by the capture;
An image conversion unit that converts a format from a 2D image to a 3D image;
An image buffer for 3D images for writing an image converted by the image conversion unit or an input 3D image;
A stereoscopic image display apparatus comprising: a display unit that displays a 3D image switched by the switching unit or an image written in the image buffer for 3D images. A device for displaying a stereoscopic image,
A 2D image buffer for writing the input 2D image;
An image conversion unit that converts a format from a 2D image to a 3D image;
An image buffer for 3D images to which the image converted by the image conversion unit or the 3D image switched by the switching unit is written;
A coordinate converter for converting the format from the input 3D image and position information;
A stereoscopic image display device comprising: a display unit that displays an image written in the image buffer for 3D images or an image that has undergone format conversion by the coordinate conversion unit. A method of displaying a stereoscopic image,
A step of switching by the switching unit whether the input image is a 2D image or a 3D image;
Acquiring 2D images switched by the switching unit by capture;
Converting the image acquired by the capture from a 2D image to a 3D image by an image conversion unit;
3D image switched by the switching unit or 3D format converted by the image conversion unit
And a step of displaying an image on a display unit.

Images