JP4598856B2 - 3D display method - Google Patents

Description

    The present invention relates to a stereoscopic display method using multi-directional display of a screen.

As a method for displaying a three-dimensional image, a method of reproducing an image of an object taken from many directions at the same time on a stereoscopic display device can be considered. However, it is physically difficult to display different images (because they are taken from many directions) from the same point in multiple directions at the same time, and many viewers (in the present invention, audio is not related, but for convenience. It is very difficult to reproduce a stereoscopic image that allows a viewer to observe a target from various directions at the same time, and to display such a stereoscopic video in real time.

In screen formation, when various elements are used, at present, the screen is generally opaque when viewed from a different direction from the display screen. If there is a screen panel that is transparent when viewed from other than the display direction, it is possible to prepare the necessary number of screens for each of the above directions and keep them always lit. The image in the viewing direction is not obstructed by the image for the other direction), and the stereoscopic image as described above can be easily realized. However, there is currently no such image panel.

Here, it is shown how to realize a stereoscopic image as described above using a current type screen that is opaque when viewed from other than the display direction. A “screen panel” is a device that temporarily generates a screen in an arbitrary direction by applying a voltage to a flat area in a transparent solid even with a single physical panel (hereinafter referred to as “ May be referred to as a “three-dimensional panel”. However, there is a limitation in a physical panel, and the method of claim 3 assumes a three-dimensional solid panel.

The first aspect of the present invention is a method of generating images by continuously switching screens in a short time, and as a result, displaying images in multiple directions. At this time, even if the display is simply switched at a high speed, it cannot be recognized as an image by human eyes, and therefore, it is continuously displayed for a certain period of time (tens of seconds) in one display in one direction. In addition, since images are displayed intermittently in multiple directions, the image may be blurred if images other than those that should be visible to the relevant viewer are mixed.Therefore, the viewing angle of the screen is set so that images for other directions do not enter. The point is to narrow it appropriately. However, the viewing angle is narrowed only in the direction in which stereoscopic vision is provided, and the viewing angle is not narrowed in other directions.

A second aspect of the present invention is a method that enables image display in more directions or frequencies when using the method of the first aspect.

A third aspect of the present invention is a method of generating a single connected image by connecting and controlling a number of stereoscopic display devices according to the method of the first aspect.

Claim 4 is not a means for solving the problem, but is a display method using the method of claim 1.

According to the first aspect of the present invention, the current opaque type screen is used to display an image in multiple directions, and a stereoscopic video or still image is displayed so that a large number of people can observe the object simultaneously from various directions. be able to.

According to the second aspect of the present invention, in the stereoscopic display method of the first aspect, it is possible to display images in a plurality of directions at each moment, so that it is possible to give a margin to the display processing, and the display duration time in one direction can be increased. By making the setting flexible or increasing the density of the display direction, it is possible to make the stereoscopic image transition more smoothly with respect to the movement of the viewer, or for 1 second when handling a movie It is possible to increase the number of frames.

According to the invention of claim 3, in the stereoscopic display method of claim 1, a stereoscopic image with a wider field of view is formed, and a stereoscopic moving image display in a compact size such as a flat surface is made possible.

According to the invention of claim 4, in the stereoscopic display, various objects can be displayed and displayed in a three-dimensional manner in one space by changing the display object according to the viewing direction. Of course, the more types of items to be displayed, the narrower the angle at which one can be viewed stereoscopically. However, depending on the application, it may be sufficient to display in this format. Even in such a case, the original original body is rotated to prepare an omnidirectional image so that the entire target can be seen by gradually changing the viewing angle while limiting the stereoscopic angle. It is also possible.

Here, a case where a physical screen panel is used will be described.

For physical screen panels, rotation is used (FIG. 1). Here, based on the method of claim 2, the two screen panels are arranged orthogonally. The display object is photographed by a camera arranged along the circumference. In the case of a still image, one camera may circulate. It seems that if you increase the number of cameras and have images taken from a number of angles, you can generate a smoother 3D image. It cannot be increased. While the screen panel is rotating, the display is switched to an image corresponding to the rotation position (an image taken from that angle) in order to realize stereoscopic viewing. Images taken from the opposite side are displayed on the front and back of the panel (FIG. 2).

The screen displayed in each direction does not need to respond only to the moving angle of the viewer. Along with the circumference other than the horizontal circumference of the target object, the camera can be placed or photographed in addition to the horizontal direction and the image forming the circumferential direction can be freely selected from them. By displaying a three-dimensional image or gradually shifting the shooting angle and the display angle, it is possible to display the entire image of the object even when the viewer stops.

Even if the display structure as described above is simply rotated at high speed, it cannot be recognized as an image by human eyes. The viewer simply sees the camera image passing at high speed. Therefore, in one display in each direction, the screen panel needs to be stopped for a moment and continuously displayed (several tens of seconds). If the duration of one display is reduced, it is possible to display in more directions (of course, it is necessary to prepare the camera for that), and it is thought that the transition of the stereoscopic image due to the movement of the viewer will be smooth. However, the reduction in the display duration time seems to have a limit in recognition with the human eye.

However, this alone still does not give stereoscopic vision. The viewer should be able to see only the image that should be seen when viewed from that direction, but if this is the case, the image on the other panel will also enter the field of view and the image will appear blurred. Therefore, it is necessary to narrow the viewing angle of the screen to an appropriate angle only in a specific direction. Basically, a place where the next prepared video is switched as the viewer moves (the boundary of the region corresponding to the next nearby shooting position) is defined as the boundary of the viewing angle. If the viewing angle is too narrow, the image is interrupted before the viewer recognizes the adjacent image, and a blank area where nothing can be seen is considered to be formed. In FIG. 4, the viewer must be able to see only the image of the camera 2 displayed when the screen panel comes to the position A. The viewing angle is narrowed only in the direction in which stereoscopic expression is possible (the image for the direction is prepared and transition is possible), and in this case, the horizontal direction is along the circumference. Since a series of images that generate stereoscopic vision cannot be displayed even if moved in the vertical direction, limiting the viewing angle is counterproductive (if it is limited, the image disappears only by changing the eye height) ) Although the viewing angle control method is not described here, it is conceivable that a physical barrier around the pixel is generated when the surface layer of the panel is formed.

In displaying moving images, it is ideal to provide images of about 30 frames / second. However, in the method of claim 1, since there is a switch in another direction, basically, the video cannot be displayed in that direction while the video is output in the other direction. For example, if the video display duration in one direction is 0.01 seconds, 0.3 seconds is spent at 30 times / sec display, so display in the other direction is 0.7 seconds (without considering switching time, etc.) Only. Although it is considered that the display can only be performed roughly in three directions in total, in practice, by using the front and back of the screen and further crossing the display screen vertically by the method of claim 2, four times (total 12). The video can be displayed in the same direction. In addition, the balance between the number of frames, the number of display directions to be used, and the video duration will be determined on a case-by-case basis depending on whether still images or moving images are important.

Although the high-speed rotation and stop of the panel may become a structural bottleneck, FIG. 5 adopts the second aspect, and when the screen panel is not rotated once, the next rotation in the same direction only needs to be rotated 90 degrees. Indicates that it can be displayed. As a result, the rotational speed can be reduced to ¼.

In this implementation method, as in the vertical direction, a stereoscopic view cannot be given to a change in the distance of the viewer in the direction of the display device.

Several examples are given below.

[Example 1]
Here, a case of a three-dimensional stereoscopic panel will be described. The three-dimensional panel is normally in a transparent state, and can generate an image only in any specific plane area at any time by applying voltage or the like (FIG. 6). In the figure, although a plurality of screens are drawn, basically, the screen that can be displayed at each moment is only “one direction” (and the direction perpendicular to the back surface) by the method of claim 1. However, as will be described later in Example 3, if a panel having such a basic structure is concatenated as a single three-dimensional panel, the screen is displayed in multiple directions at each point in time. I can say that.

In the case of a three-dimensional panel, the physical mechanism load that temporarily stops the rotation does not occur compared to the method of rotating the physical panel described above. Therefore, the display direction of the screen is free, and in principle, it is possible to give a stereoscopic view to any movement along the spherical surface. The photographing camera is not arranged on the circumference, but is freely arranged three-dimensionally with respect to the display target. It is also possible to divide and display a screen in one direction into a number of different screens, and there is a degree of freedom in size and shape depending on the application. Further, if it is not necessary to stereoscopically view all 360 degrees around depending on the purpose of use, the amount of display processing in that direction can be used to increase the display direction density toward the stereoscopic viewing side.

Viewing angle control in a three-dimensional 3D panel will allow free viewing angle control in multiple directions (assuming stereoscopic viewing in both horizontal and vertical directions). It will be controlled by application (barriers at necessary positions surrounding the pixels are generated and extinguished by applying voltage, etc.).

Even in this embodiment, the stereoscopic view is not given to the change in the distance to the display device.

[Example 2]
An embodiment of claim 3. A number of stereoscopic display devices according to the method of claim 1 are connected in a two-dimensional direction. In FIG. 8, they are arranged in a straight line as an example. As is clear from the screen position shown in this figure, this is not possible with a physical panel having a fixed axis of rotation, so it is assumed that a three-dimensional solid panel is used.

In this method, a stereoscopic image with a wide field of view can be provided to the viewer, and the apparatus size can be made compact compared to a case where an image in one direction is displayed on a single screen. However, it does not provide a stereoscopic view when wrapping around the side or back of the device (a stereoscopic image can be displayed independently on the back of the device). The viewer obtains a stereoscopic view by moving in parallel with the connection direction.

Each one device plays a role like a cell (hereinafter referred to as a cell). A cell is considered to be a unit in which adjacent areas are grouped to display a screen in the same direction (in some cases, a vertical screen in some cases) at each moment (that is, a unidirectional screen is composed of a plurality of screens). May be good).

Although FIG. 8 shows only the connection in the horizontal direction, it is also possible to connect cells in the vertical direction to give a stereoscopic view in the vertical direction. The shooting points spread two-dimensionally, and if it is a moving image, only that many cameras are prepared (FIG. 7). In addition, each camera needs to have a wide-angle or super-wide-angle so that all images necessary for generating a stereoscopic view can be obtained. The direction in which the viewer can stereoscopically view by movement is basically the camera arrangement direction, which is a direction parallel to the connection surface.

In FIG. 8, note that the screen display angle for the viewer in the figure differs for each cell. In addition, the images of adjacent cells overlap, and there are invalid portions that are not visible to the viewer (displayed in cell C). This part may or may not be displayed, but can be used for display in other directions (viewers in other positions) and used to earn display counts in other directions.

If the depth of the device is further reduced and the back surface of the cell is brought up to the position A in FIG. 8, one screen in the cell cannot fit within the cell in size. In that case, the corresponding screen is formed with a plurality of screens. This is shown in cell B in the figure.

As one control method for connected cells, each cell manages only the type and direction of the screen that it displays, and does not link the timing with other cells. Within a single cell, the cycle of displaying an appropriate image in various directions is repeated at high speed, but this method maintains continuity as a stereoscopic image between cells, but The display timing in each direction in the cycle is not synchronized. That is, at a certain moment, the cell A is displaying an image for the viewer in the figure, but the adjacent cell B may be in the middle of displaying an image for another viewer position. (In the figure, the screen in the cell is synchronized with one viewer direction)

Example 3
An embodiment of claim 4. Change the viewing direction and the display target. A plurality of objects are captured by a camera in real time or recorded. This is easy because the video source is simply changed depending on the screen position.

The conceptual diagram from the video source for producing | generating a stereo image at the time of using a physical screen panel to a stereo image production | generation apparatus. It is a figure which shows changing the image displayed with rotation of a screen (in the case of physical screen panel rotation). It is a figure which shows that the screen display to a multi-direction can be performed in each moment with the method of Claim 2. It is a figure which shows the necessity to narrow a viewing angle in a stereoscopic vision direction. It is a figure which shows that the next image | video can be taken out by 90 degree rotation by the method of Claim 2 (in the case of physical screen panel rotation). It is a figure which shows the case where a three-dimensional solid panel is used. It is a figure which shows taking in of the target object image | video in the case of implementing the method of Claim 3. It is a figure which shows the coupling | bonding of a cell when implementing the method of Claim 3, and the screen inside a cell.

Explanation of symbols

Claims (1)

By forming a barrier in the pixel surrounding the 3D panel and creating and extinguishing it by applying voltage, the viewing angle can be set at any location and in any direction within the 3D panel. To display images

Images