JP2010014891A - Three-dimensional image display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problems that in a conventional multi-view 3D display device, remarkable discontinuity of a 3D image and switching of images are generated when a viewer moves his or her line of sight and the image gives a strange feeling to the viewer as a continuous 3D object, and according to the conditions for presenting the 3D video image, more remarkable discontinuity and switching of the image are generated as a deviation (amount of pop-up) from a display screen is increased, thereby giving a strange feeling. <P>SOLUTION: In this device for reproducing a multi-view 3D video image based on three or more pieces of parallax image information, a region for a crosstalk image with the amount of crosstalk corresponding to the deviation of a 3D object to be presented is provided when the viewer views the multi-view 3D video image, and the 3D image is presented by an image separation width of a parallax image corresponding to the deviation, to provide the 3D video image for preventing discontinuity and switching of the images. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a space-divided multi-view three-dimensional display that can display a continuous three-dimensional image without making the switching and image skip of multi-view three-dimensional images observed by an observer's horizontal head movement inconspicuous.

  As a conventional multi-viewpoint image display device that expresses motion parallax, there is a parallax / panoramagram method.

  FIG. 11 shows an example of a parallax / panoramagram multi-viewpoint image display apparatus 1001, in which eight different stripe images 1 to 8 are horizontally separated from the multi-viewpoint image display apparatus 1001 to an observation position 400 at a distance L0. Can be displayed.

  A multi-viewpoint image display apparatus 1001 includes a display device 201 and a mask 301.

  The display device 201 includes a CRT, a liquid crystal display, a plasma display, and the like, and displays a stripe composite parallax image (hereinafter referred to as a composite image) created using images taken from eight different positions.

  On the display surface side of the display device 201, a mask 301 is provided in which a plurality of vertical slit-shaped openings 311 and light shielding 321 are arranged in the horizontal direction.

  12 and 13 illustrate the input and composition of multi-viewpoint images.

  The subject 401 in FIG. 12 is input using a compound-eye camera 501 provided with eight cameras 511 to 518 arranged at equal intervals e in the horizontal direction. FIG. 13 shows eight images 611 to 618 captured by the cameras 511 to 518, respectively. Images 611 to 618 taken from the eight different cameras are divided into vertically long strips, and these strip images are sequentially repeatedly arranged in the horizontal direction and combined to create a composite parallax image 601.

  FIG. 10 is a horizontal sectional view of the multi-viewpoint image display device 1001.

  The observer can observe a natural stereoscopic image having continuous motion parallax according to the movement of the horizontal viewpoint (observation position) in the region of the width 8e at the distance L0 from the multi-viewpoint image display device 1001. .

  The region e becomes the observation amount region of the parallax image for one viewpoint, and when the observer moves horizontally, the region e enters the viewing region of the parallax image of the adjacent viewpoint. Become.

As an example of the above, Patent Document 1 below can be cited.
JP 2006-106608 A

  By the way, in the multi-viewpoint image display device of the above-described conventional example, motion parallax is reproduced, but since the parallax is discretized by a large number of images, it is necessary to always move the eye position while moving the image separation width in the horizontal direction. Once you perceive the transition of the video.

  FIG. 6 is an explanatory diagram of a luminance distribution waveform on the observation surface 400 of the multi-viewpoint image display device 1001 of FIG. 601 to 605 in FIG. 6 are triangular waveforms as an example of the luminance distribution waveform on the observation surface 400 that has passed through the opening width 311.

  The observation eye 501 can observe the parallax image corresponding to the viewpoint 1 with the maximum luminance, and the observation eye 503 can also observe the parallax image corresponding to the viewpoint 2 with the maximum luminance that is the vertex of the luminance waveform (triangular wave). However, the observation eye 502 becomes a boundary point between the adjacent viewpoint 1 and viewpoint 2, and the parallax image is observed with the minimum value of the luminance waveform. Sometimes, the parallax image may appear black at this boundary point.

  The passing of the observation eye on such a luminance waveform causes the inconvenience that the switching of the image is perceived through the change in luminance.

  7 shows the difference in the parallax image (parallax amount) that is the basis of the composite image displayed on the display surface 201 when the multi-viewpoint image display device 1001 in FIG. It is a figure explaining.

When it is assumed that the observation object 501 corresponding to the viewpoint 1 and the image separation width e move in the horizontal direction at the observation distance L0 and the three-dimensional object is observed with the two eyes 503 corresponding to the viewpoint 2, the display display surface 350 The parallax images constituting the three-dimensional object B1 with the deviation amount B10 are B11 and B12, and the parallax amount between B11 and B12 is B13.
Further, the parallax images constituting the three-dimensional object B2 with the deviation amount B20 are B21 and B22, and the parallax amount between B21 and B22 is B23.

  In this way, the disparity constituting the composite image displayed on the display surface 201 becomes larger as the deviation amount (protruding amount) of the stereoscopic video presented on the multi-viewpoint image display apparatus 1001 deviates from the display surface (the popping amount is large). The amount of parallax of the image increases (B13 <B23).

  For this reason, as the amount of parallax increases, the amount of change in the image increases with respect to the amount of movement in the horizontal direction. This is perceived as “image flipping”, and as a result, perceived as a continuous stereoscopic image. The inconvenience that it is not done arises.

  The present invention eliminates the problems of the above-mentioned conventional example, makes the switching of the multi-view stereoscopic video perceived by moving the position of the observer's eye left and right by the observer's horizontal head movement, and makes the image jump inconspicuous, continuously. An object of the present invention is to provide a multi-view stereoscopic display capable of displaying a stereoscopic image.

The three-dimensional image display device according to the present invention is
In a three-dimensional image display device that receives parallax image information from N viewpoints (N is a natural number of 3 or more) as input, and presents parallax images to viewers independently in M (M is a natural number of 3 or more) different directions.
It has a means to change the amount of crosstalk between parallax images according to the amount of divergence between the presentation position of the stereoscopic image to be presented and the display surface.

  In particular, the present invention is characterized by having means for monotonously reducing the amount of crosstalk between parallax images as the amount of deviation between the presentation position of the stereoscopic image to be presented and the display surface increases.

  Further, the present invention is characterized by having means for changing the image separation width of the parallax image in accordance with the amount of deviation between the presentation position of the stereoscopic image to be presented and the display surface under the condition of N> M.

  In particular, it is characterized by having means for monotonously reducing the image separation width of the parallax image in accordance with the amount of deviation between the presentation position of the stereoscopic image to be presented and the display surface under the condition of N> M.

  Further, the three-dimensional image display device is characterized by having means for judging a pop-out amount / presentation range of a stereoscopic image presented by an observer and inputting the pop-out amount.

  In the three-dimensional image display device, the pop-up amount of the stereoscopic image to be presented is read, and the pop-out amount is detected by reading the presentation position information recorded in the header area of the parallax image information constituting the stereo image. Yes.

  In the three-dimensional image display device, the pop-out amount of the stereoscopic image to be presented is detected from the correlation between the parallax image information constituting the stereo image.

  In an apparatus that reproduces multi-view stereoscopic video based on three or more pieces of parallax image information, when an observer views multi-view stereoscopic video, crossing with a crosstalk amount corresponding to the amount of deviation of the three-dimensional object presented By providing a talk image appreciation area and presenting it with the image separation width of the parallax image corresponding to the amount of deviation, it is possible to provide a 3D image in which image skipping and image switching are not perceived.

  Next, details of the present invention will be described in accordance with the description of the embodiments.

  FIG. 1 is a schematic view showing a first embodiment of the apparatus.

  In the figure, reference numeral 201 denotes a display, which is a liquid crystal element (LCD) having a backlight light source, for example, and its display surface is composed of a large number of pixels having a matrix structure, and displays an image with non-interlaced scanning lines. 1 to 8 are schematic views schematically showing a state of a stripe parallax image described later displayed on the pixel display surface of the display 201. 450 is a diagram schematically showing the stripe composite images 1 to 8 in which the stripe parallax images 1 to 8 displayed on the display 201 pass through 301 described later and are observed on the observation surface 400.

  Reference numeral 301 denotes a spatial light modulation element, which is composed of a transmissive liquid crystal element or the like, and its display surface is composed of a large number of pixels having a matrix structure. When displaying a stripe parallax image on the display 201, a predetermined pitch is used. The light transmission part (opening part) 311 and the light shielding part 321 are arranged in the horizontal direction to form (display) a parallax barrier pattern (light-opening pattern).

  A1 is a parallax image source of N-system original parallax images that is a basis of a stripe parallax image described later, A2 is a three-dimensional object deviation input means, A3 is a width of the opening 311 of the spatial light modulator 301, An aperture width / display width determining means for determining the image separation width e of the stripe composite image observed on the observation surface 400, A4 is an image processing means for a stripe parallax image described later, and A5 is a barrier drive for driving the spatial light modulator. A circuit A 6 is a display driving circuit for driving the display 201.

  The parallax image source A1 is the same as that shown in FIG. 12 described in the above-described prior art, and includes, for example, a VTR input from multi-channel cameras 511 to 518, or three-dimensional data of a subject. Hereinafter, a plurality of future images and three-dimensional data will be referred to as parallax image information. Note that a multi-channel VTR, a multi-channel imaging device, and the like have a plurality of images, but a parallax image (an image having a parallax) is selected from these images, so that these plurality of images are converted into an original parallax image. I will call it.

  The three-dimensional object divergence amount input means A2 uses a display driving circuit A6 and a barrier driving circuit A5, which will be described later, to observe parallax image information that the observer wants to observe as a three-dimensional image on the observation plane 400, or This is a means for determining and inputting the amount of deviation from the display based on the subject most distant from the display surface 350. The input means is a means for inputting the amount of deviation by an input device composed of a plurality of buttons, for example, ± 0.05D (D: diopter is the reciprocal of meter), ± 0.10D based on the display surface Means for the observer to judge and input using an input device composed of buttons corresponding to a plurality of deviation conditions such as ± 0.15D can be considered.

  FIG. 2 is a flowchart for explaining the opening width / display width determining means A3. Based on the deviation amount of the stereoscopic image selected from the deviation amount input means A2 from the display surface 350, the optimum mask opening width 311 is shown. Then, the image separation width e of the stripe composite image is determined.

  When the deviation amount J is inputted from the deviation amount input means A2, the opening width G is transmitted to the barrier driving means A5 because the deviation amount J satisfies the deviation condition F1. In addition, the image separation width D is transmitted to the image composition means A4. When the divergence amount K is input, the opening width H is transmitted to the Mavalier driving means A5 and the image separation width E is transmitted to the image processing means A4 in order to satisfy the divergence condition F2. When the deviation amount L is input, the opening width I is transmitted to the barrier driving means to satisfy the deviation condition F3, and in addition, the image separation width F is transmitted to the image composition means.

  FIG. 8 is a graph showing the relationship between the deviation amount and the opening width 311 of the mask, and is an example showing the relationship that the opening width of the mask monotonously decreases as the deviation amount increases.

  FIG. 9 is a graph showing the relationship between the deviation amount and the image separation width, and is an example showing the relationship that the image separation width e of the stripe composite image on the observation surface monotonously decreases as the deviation amount increases.

  A5 is a barrier drive circuit, which drives the spatial light modulator 301 by an aperture width signal from the aperture width / display width determining means A3 and forms a parallax barrier pattern thereon. Hereinafter, the effect when the opening width is changed will be described.

  FIG. 3 shows a luminance distribution waveform on the observation surface where the stripe parallax image displayed on the display device 201 has passed through the spatial light modulator 301 having the aperture width O1 (O1≈0).

  An interval R2 between the points T1 and T3 where lines S1 and S2 extending from the both ends of the stripe parallax image 2 through the opening width O1 to the observation surface 400 intersect at the observation position 400 is an observation region of the viewpoint 2. A point T2 where the observation position 400 intersects with a dotted line TS1 extending from the center of the stripe parallax image 2 through the opening width O1 to the observation position 400 is the center of the observation region R2 of the viewpoint 2.

  Further, an interval R3 between the point T3 and the point T5 at which the lines S2 and S3 extending from the both ends of the stripe display image 3 through the opening width O1 to the observation surface 400 intersect at the observation position 400 is an observation region of the viewpoint 3. The point T4 where the dotted line TS2 extending from the center of the stripe display image 3 through the opening width O1 to the observation position 400 and the observation position 400 intersect is the center of the observation region R3 of the viewpoint 3.

  W2 is a luminance distribution waveform on the observation surface of the observation region R2 of the viewpoint 2, and W3 is a luminance distribution waveform on the observation surface of the observation region R3 of the viewpoint 3.

  The luminance at the boundary point T3 between the observation area R2 of the viewpoint 2 and the observation area R3 of the viewpoint 3 is the minimum value, and the center point T2 of the viewpoint 2 and the center point T4 of the viewpoint 3 are the maximum value.

  FIG. 4 shows a luminance distribution waveform on the observation surface with an opening width O2 (O1 <O2).

  An interval R2 between a point T1 and a point T31 where lines S1 and S2 extending from both ends of the stripe parallax image 2 through both ends of the opening width O2 to the observation surface 400 intersect at the observation position 400 is an observation region of the viewpoint 2. A point T2 where the dotted line TS1 extending from the center of the stripe display image 2 through both ends of the opening width O2 to the observation position 400 and the observation position 400 intersect is the center of the observation region R2 of the viewpoint 2.

  In addition, an interval R3 between the points T51 and T32 where lines S31 and S32 extending from both ends of the stripe display image 3 through both ends of the opening width O2 to the observation surface 400 intersect at the observation position 400 is an observation region of the viewpoint 3. . A point T4 where the observation position 400 intersects with a dotted line TS2 extending from the center of the stripe display image 3 through both ends of the opening width O2 to the observation position 400 is the center of the observation region R3 of the viewpoint 3.

  W2 is a luminance distribution waveform on the observation surface of the observation region R2 of the viewpoint 2, and W3 is a luminance distribution waveform on the observation surface of the observation region R3 of the viewpoint 3.

  CT is an area (crosstalk area) where the observation area R2 of the viewpoint 2 and the observation area R3 of the viewpoint 3 overlap. Therefore, in the crosstalk region CT, the luminance distribution waveform W2 of the viewpoint 2 and the luminance distribution waveform W3 of the viewpoint 3 are added, and the central point T2 of the viewpoint 2 and the maximum luminance of the central point T5 of the viewpoint 3 are The luminance difference Δ is smaller than that at the opening width O1.

  A4 is an image processing means, which is a processing means for synthesizing a parallax image as shown in FIG. The display width H1 of the parallax image on the display device 201 is calculated from the optimum image separation width e obtained from the aperture width / display width determination means A5, and all the parallax images are horizontally calculated with the calculated display width. This is a means for dividing and generating vertically striped stripe parallax images, which are alternately arranged and combined into a single stripe image.

  An example of a formula for calculating the display width H1 of the parallax image from the aperture width / display width determination unit A5 will be described with reference to FIG. The image separation width e is equal to the image viewing area R2 of the viewpoint 2 and the image viewing area R3 of the viewpoint 3, and the viewing position 400, the display device 201, and the mask 301 are in a parallel relationship, and thus the following relationship is established.

R2: L0 = H1: L1
Since L1 and L0 are determined by the design value of the display, and R2 that is the image separation width e is obtained from the aperture width / display width determination means A5, H1 is calculated by the following equation.

H1 = R2 * L1 / L0
A6 is a display driving circuit, and displays the stripe composite image 601 of FIG. 13 which is synthesized and output by the image processing means A4 on the display surface of the display.

  The relationship between the stripe composite image 601 and the parallax barrier pattern of this embodiment will be described with reference to FIG. The distance between the eyes (baseline length) of the observer is PD, the observation distance is L0, the distance between the display device 201 and the spatial light modulator (parallax barrier) 301 is L1, and the parallax Assuming that the width of the opening of the barrier pattern is O1, and the display width of the stripe pixels constituting the stripe composite image displayed on the display device 201 is H1, the following relationship is satisfied between them in order to obtain a stereoscopic view. There is a need.

・ L1 = H1 * L0 / (PD + H1)
・ O1 = H1 * (L0-L1) / L0
Actually, the observation width at the observation position 400 has a finite width, and these various amounts are set with some changes.

  FIG. 5 is a schematic view showing a second embodiment of the apparatus.

  Other than the A11 parallax image source, the description is omitted because it is basically the same as the first embodiment described above.

  In addition to the parallax image source A1 of the first embodiment, the parallax image source A11 includes the focal length information of the camera set by the photographer with the cameras 511 to 518 in FIG. Alternatively, the projection amount (deviation amount) data of the difference in the focal length is recorded in the header area of the image format of the parallax image source, and this data is input to the aperture width / display width determination means A3 as the deviation amount. It is.

  Alternatively, the correlation coefficient between the same scanning lines of the parallax images input from two cameras of the compound eye camera is calculated, and it is determined that the larger the correlation coefficient is, the larger the divergence amount (protrusion amount) is. The deviation amount is input to the opening width / display width determination means A3.

Schematic showing the first embodiment of the apparatus Judgment means for aperture width / image separation width Schematic showing the mask aperture width (small) and the luminance distribution waveform on the observation surface Schematic showing the aperture width (large) of the mask and the luminance distribution waveform on the observation surface Schematic showing a second embodiment of the apparatus Diagram showing the luminance distribution waveform on the observation surface The figure which showed the difference in the amount of parallax of the solid thing from which the amount of deviation differs Diagram showing the relationship between the amount of deviation and the opening width of the mask Diagram showing the relationship between the amount of deviation and the image separation width Horizontal view of multi-view display device Schematic diagram of a parallax panoramagram multi-viewpoint display device Explanatory drawing about input of parallax image Illustration of a method for synthesizing parallax images

Explanation of symbols

1-8 stripe parallax image (on 201) and stripe composite image (on 400)
DESCRIPTION OF SYMBOLS 201 Display device 301 Mask 311 Mask opening part 321 Mask light shielding part 350 Display surface of display 1001 Multi-viewpoint image display device L0 Observation distance L1 Distance between display device and mask 400 Observation position 450 Viewing surface 401 Subject e Image separation width 511- 518 Camera 501 Compound eye camera 611-618 Photographed image of camera 511-518 601 Stripe composite image 601-605 Luminance distribution waveform 501-503 Observation eye B1 Three-dimensional object of divergence amount B10 B11 Parallax image of divergence amount B10 B12 Parallax of divergence amount B10 Image B13 Parallax amount of divergence amount B10 B10 Deviation amount B2 Three-dimensional object of divergence amount B20 B21 Parallax image of divergence amount B20 B22 Parallax image of divergence amount B20 B23 Parallax amount of divergence amount B20 B20 Divergence amount A1 Parallax image saw of Embodiment 1 A11 Parallax image source of Embodiment A2 Solid object deviation amount input means A3 Aperture width / display width judgment means A4 Image processing means A5 Barrier drive circuit A6 Display drive circuit F1 Deviation quantity judgment condition F2 Deviation quantity judgment condition F3 Judgment condition for deviation amount D Large image separation width E Medium image separation width Small F Image separation width Small G Aperture width Large H Aperture width Medium H1 Display width I Small opening width J Small amount of divergence Small K Large amount of divergence Large S1 Ray S2 Ray S3 Ray S4 Ray TS1 Ray TS2 Ray O1 Aperture width small O2 Aperture width large T1 Intersection of the ray and the observation position T2 Intersection of the ray and the observation position T3 Intersection of the ray and the observation position T31 Intersection of the ray and the observation position T32 Intersection of ray and observation position T4 Intersection of ray and observation position T5 Intersection of ray and observation position TS1 ray TS2 ray CT cross Over click region R2 image separation width R3 image separation width W2 luminance distribution W3 luminance distribution

Claims (7)

  Display parallax image information from N viewpoints (N is a natural number of 3 or more) as input, and present stereoscopic images to be presented to the viewer independently in different directions of M (M is a natural number of 3 or more). A three-dimensional image display device comprising means for changing a crosstalk amount between parallax images according to a deviation amount between a position and a display surface.   Display parallax image information from N viewpoints (N is a natural number of 3 or more) as input, and present stereoscopic images to be presented to the viewer independently in different directions of M (M is a natural number of 3 or more). 2. The three-dimensional image display device according to claim 1, further comprising means for monotonously decreasing the amount of crosstalk between parallax images as the amount of deviation between the position and the display surface increases.   With parallax image information from N viewpoints (N is a natural number of 3 or more) as input, parallax images are independently presented to viewers in M (M is a natural number of 3 or more) different directions, and N> M A three-dimensional image display device comprising means for changing an image separation width of a parallax image in accordance with a deviation amount between a presentation position of a stereoscopic image to be presented and a display surface.   With parallax image information from N viewpoints (N is a natural number of 3 or more) as input, parallax images are independently presented to viewers in M (M is a natural number of 3 or more) different directions, and N> M 4. The three-dimensional image display device according to claim 3, further comprising means for monotonously decreasing the image separation width of the parallax image in accordance with the amount of deviation between the presentation position of the stereoscopic image to be presented and the display surface.   The parallax image information from N viewpoints (N is a natural number of 3 or more) is input, the parallax images are presented to the viewer independently in different directions of M (M is a natural number of 3 or more), and the stereoscopic image presented by the viewer The three-dimensional image display device according to any one of claims 1 to 4, further comprising means for determining a pop-out amount / presentation range of an image and inputting the pop-out amount.   The parallax image information from N viewpoints (N is a natural number of 3 or more) is input, and the parallax image is presented to the viewer independently in different directions of M (M is a natural number of 3 or more). The three-dimensional according to any one of claims 1 to 4, wherein the amount of projection is read out from the presentation position information recorded in the header area of the parallax image information constituting the stereoscopic image, and the amount of pop-up is detected. Image display device.   The parallax image information from N viewpoints (N is a natural number of 3 or more) is input, and the parallax image is presented to the viewer independently in different directions of M (M is a natural number of 3 or more). The three-dimensional image display device according to any one of claims 1 to 4, wherein the amount of protrusion is detected from a correlation coefficient between pieces of parallax image information constituting a stereoscopic image.

Images