KR20110093616A - Method for converting overlaying area into 3d image and apparatus thereof - Google Patents

Abstract

PURPOSE: A three-dimensional image converting method of an overlay area and apparatus thereof are provided to reduce audition fatigue according to 3D effect of an overlay region. CONSTITUTION: A depth map indicating perspective is generated from an inputted mono-scopic image(210). The overlay region included in the inputted mono-scopic image is detected(220). The depth value of a map is corrected in the detected overlay area(230).

Description

Method for converting 3D image of overlay area and apparatus therefor {Method for converting overlaying area into 3D image and apparatus}

The present invention relates to a 3D image conversion method of the overlay region, and more particularly, to a 3D image conversion method of preventing distortion by making the overlay region have a uniform depth in the conversion to the 3D image.

The biggest factor that humans feel in 3D is due to the spatial difference on the left and right retinas caused by looking at one object from different directions to the left and right eyes. In order to be able to feel such a difference in the display device which is a two-dimensional plane, a viewer may feel that the user is viewing a three-dimensional image by displaying different images, that is, a stereoscopic image.

Such a 3D image may be generated by using a plurality of cameras at the time of production, but may be converted into a 3D image later on. In this 3D image conversion technology, the left and right eye images are separated and generated in consideration of binocular disparity, and then combined and played back, and the viewer separates the image from the left eye and the image from the right eye by wearing a shutter and polarized glasses. By observing, it is felt to see three-dimensional images.

However, in the conventional 3D image conversion technology of the subtitle area, there is no separate processing method for the subtitle area included in the 2D image. However, when the metadata for the subtitle is received in addition to the image data, Data can also be used to coerce subtitles into 2D.

However, since the subtitle area is forcibly converted to 2D, after the 3D image conversion, a viewing disturbance occurs due to the discontinuous depth of the subtitle area and the surrounding background, or the subtitle area is displayed behind the background image. This leaves the viewing fatigue still.

An object of the present invention is to uniformly allocate the depth value to the overlay area in consideration of the relationship with the depth of the surrounding background for the area after detecting the overlay area for the general 2D image having no metadata for the overlay area Accordingly, the present invention provides a method and apparatus for converting 3D images of an overlay area to reduce viewing fatigue due to the 3D effect of the overlay area and to prevent discontinuity of the overlay area and the surrounding background from occurring.

According to an aspect of the present invention, there is provided a method of converting a 3D image of an overlay region, the method including: generating a depth map displaying a perspective from an input monocular image; Detecting an overlay region included in the input monocular image; Correcting a depth value in the depth map with respect to at least one of the detected overlay area and the background image; Generating a binocular image based on the corrected depth map.

According to another embodiment of the present invention, the correcting step is corrected to a value close to the depth value assigned to the peripheral area of the detected overlay area.

According to another embodiment of the present invention, the correcting may include searching for at least one depth value with respect to a background image located at a boundary of the detected overlay area; Calculating an average value of the retrieved at least one depth value; And uniformly allocating the calculated average value as a depth value of the overlay area.

According to another embodiment of the present invention, the detecting of the overlay region may include extracting an overlay region by analyzing edges and color components in the input monocular image.

According to another embodiment of the present invention, generating the binocular image generates the binocular images of the left and right eyes by calculating the binocular disparity from the completed depth map by reflecting the correction value for the overlay area.

According to another embodiment of the present invention, the detecting of the overlay area is performed based on the location information of the overlay area, and the location information of the overlay area is a coordinate representing the location of the overlay area or the location of the overlay area. Bitmap information indicating a location.

According to another embodiment of the present invention, the step of correcting further comprises the step of determining whether to perform the correction based on at least one of the position, size and movement of the overlay area, The step is performed only when it is determined that the correction is performed as a result of the determination.

According to another embodiment of the present invention, in the correcting, the objects of the background image are corrected to have the same depth value.

According to another embodiment of the present invention, the correcting may include correcting such that a difference between depth values of the objects of the background image is reduced.

According to another embodiment of the present invention, the correcting may include a depth value higher by a predetermined depth than the highest depth value of the depth values of the objects of the background image or an average value of the depth values of the objects of the background image. Is set as the depth value of the overlay area.

According to another embodiment of the present invention, the correcting step sets the depth value at the center of the overlay area or the depth value at the boundary of the overlay area to the depth value of the entire overlay area.

According to another embodiment of the present invention, after the correction is performed, the method may further include performing filtering on the overlay area to remove a difference between the depth value of the overlay area and the depth value of the background image. .

According to another embodiment of the present invention, the correcting is performed using a depth value obtained by applying weights to the depth value of the background image and the depth value of the overlay area, respectively.

According to another exemplary embodiment of the present disclosure, the correcting may include one of a minimum value, a maximum value, and an average value of depth values of each of the plurality of overlay regions when the overlay regions are plural. Set to the depth value.

According to another embodiment of the present invention, the correcting step sets the depth value input from the user to the depth value of the overlay area.

According to another embodiment of the present invention, the step of generating the binocular image is performed by referring to only pixel values in the background image without referring to pixel values in the overlay area.

On the other hand, 3D image conversion apparatus of the overlay area according to another embodiment of the present invention for solving the above technical problem and a depth map generator for generating a depth map (display depth) from the input monocular image; An overlay region detector for detecting an overlay region included in the input monocular image; A depth value corrector for correcting a depth value in the depth map with respect to at least one of the detected overlay area and the background image; And a binocular image generator configured to generate a binocular image based on the corrected depth map.

According to another embodiment of the present invention, the depth value corrector corrects the depth value of the overlay area to a value close to the depth value allocated to the peripheral area of the detected overlay area.

According to another embodiment of the present invention, the depth value corrector comprises: a searcher for searching for at least one or more depth values with respect to a background image located at a boundary of the detected overlay area; The apparatus may further include an average calculator configured to calculate an average value of the at least one searched depth value, and uniformly allocate the calculated average value as a depth value of the overlay area.

According to another embodiment of the present invention, the overlay region detector extracts an overlay region by analyzing edges and color components in the input monocular image.

According to another embodiment of the present invention, the binocular image generation unit generates binocular images of left and right eyes by calculating binocular disparity from the completed depth map by reflecting the correction value for the overlay area.

In order to solve the above technical problem, an embodiment of the present invention provides a computer-readable recording medium recording a program for executing the above-described 3D image conversion method.

According to the 3D image conversion method of the overlay region according to an embodiment of the present invention, as a result of the conversion to the 3D image, even if the same letters are given different binocular disparity depending on the position, the overlay region 3D conversion proposed by the present invention Applying the technique gives a uniform binocular parallax to the overlay area, so that the overlay area has the same depth and no distortion occurs. In addition, since the overlay region has a depth similar to that of the background region positioned around the overlay region, and when the overlay region is detected, the depth value of the background image is corrected, thereby reducing the viewing fatigue due to the 3D effect of the overlay region.

1 is a diagram illustrating an example of a monocular image frame including a caption area.
2 is a flowchart for describing a 3D image conversion method of an overlay area according to an embodiment of the present invention.
3 is a flowchart illustrating a 3D image conversion method of an overlay area according to a second embodiment of the present invention.
FIG. 4 is a flowchart for describing an exemplary embodiment of correcting a depth value of an overlay region in FIG. 2.
5A is a diagram illustrating an example in which an embodiment of the present invention is not applied when converting an overlay region into a 3D image.
5B is a diagram illustrating an example of applying an embodiment of the present invention when converting an overlay region into a 3D image.
6 is a functional block diagram illustrating a 3D image converting apparatus of an overlay area according to another embodiment of the present invention.
Corresponding reference numerals in the several drawings indicate corresponding parts. Although the drawings show embodiments of the invention, the drawings are not to scale and certain features may be exaggerated to better illustrate and explain the invention.

DETAILED DESCRIPTION In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings which illustrate preferred embodiments of the present invention and the contents described in the drawings.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention.

1 is a diagram illustrating an example of a monocular image frame including a caption area.

Referring to FIG. 1, when referring to the caption area 120 viewed with interest in the present invention, other parts of the image frame 110 other than the caption may be viewed as a background. Several subjects (objects) may appear in the background, and when they are converted into 3D images, they are given different depth values, and accordingly different binocular parallaxes are displayed and displayed as 3D images.

In this case, according to the related art, when the caption region 120 is converted together when the 3D image is converted, different depth values are assigned to each letter, and different binocular disparity values may be applied. On the other hand, even in the case of only 2D conversion of the subtitle area 120 separately, since the binocular disparity value is simply set to "0" (zero), regardless of the depth of the surrounding background, it is placed in front of or based on the z-axis for perspective. It happens behind the scenes.

In this case, the viewer may feel tired when watching the image for a long time due to the 3D effect of the subtitles having a discontinuous depth feeling.

Accordingly, in the present invention, after detecting an overlaying area including a caption area, an OSD (On Screen Display) area, a broadcaster logo area, and the like for a 2D image, the present invention considers a relationship with a depth value of a surrounding background. By uniformly allocating the depth value to the corresponding overlay area, we propose a method to prevent the discontinuous depth of the overlay area and the surrounding background from occurring.

Also, when an overlay region is detected in a 2D image, a method of correcting a depth value of a background image when converting a 2D image into a 3D image is proposed.

2 is a flowchart for describing a 3D image conversion method of an overlay area according to an embodiment of the present invention.

Referring to FIG. 2, the method for converting a 3D image of an overlay region according to an embodiment may include generating a depth map indicating a perspective from an input monocular image (210) and an overlay included in the input monocular image. Detecting the area 220 and correcting the depth value in the depth map to a uniform value for the detected overlay area 230 and generating a binocular image based on the corrected depth map ( Contains 240)

First, in step 210, a depth map of the input 2D image is generated. The depth map is a map representing a three-dimensional distance difference between objects in the image and is expressed as a value between 0 and 255 for each pixel. When expressed in black / white, black (low value) indicates far away from the viewer, and white (high value) indicates near the viewer.

Next, in step 220, an overlay area included in the monocular image is detected. As an example of a technique for detecting an overlay region, there is a method of extracting an overlay region by analyzing edge and color components in a monocular image. In this case, a rectangle corresponding to the overlay area may be extracted.

In addition, overlay area extraction method is based on low-level features such as edge and color as well as texture information, or it is used to increase the height of inserted text. In the part, there is a transition region search technique using a different color from the corresponding letter. Next, in step 230, the depth value in the depth map is corrected for the detected overlay area. At this time, the correction value for the overlay area is uniformly assigned to a constant value. That is, a uniform depth value is given to the overlay area so that the sense of depth for each letter does not change.

In the case where the depth value of the overlay area is corrected as described above, the reference values used as the correction values are the depth values of the peripheral area. That is, in the converted 3D image, the caption is not heterogeneous with the surrounding background image and needs to have a value similar to the depth value of the surrounding background image so as not to be discontinuously expressed. Therefore, in one embodiment of the present invention, the depth value of the overlay area is corrected to a value close to the depth value assigned to the detected peripheral area of the overlay area.

Finally, in step 240, the left and right eyes of both eyes are generated based on the depth map completed by the correction process. That is, binocular disparity is calculated with reference to the depth map, and left and right eye images are generated by moving objects displayed in the image by each binocular disparity.

In this converted 3D image, the overlay region in the image maintains continuity by having a similar sense of depth to the surrounding background, and at the same time, each letter in the subtitle is given the same binocular disparity value so that no distortion occurs.

3 is a flowchart illustrating a 3D image conversion method of an overlay area according to a second embodiment of the present invention.

In operation 310, a depth map of the input 2D image is generated. In operation 320, the overlay area included in the monocular image is detected.

In this case, the overlay area may be detected based on the location information of the overlay area. For example, when the overlay area to be detected is an area indicating an OSD generated by the display device, the OSD area may be detected by receiving location information of the OSD area from the display device. In this case, the position information of the OSD region may have a form of coordinate values for the OSD region, but the OSD region may be displayed as 1, and the background image other than the OSD region may have a bitmap form.

In operation 330, it is determined whether at least one of the detected overlay region and the background image is to correct the depth value in the depth map.

If the depth value is corrected, go to step 340; otherwise, go to step 350.

As described above, unlike the embodiment of FIG. 2, in the present embodiment, the depth value in the depth map is not corrected for the overlay area unconditionally, but only when the overlay area satisfies a predetermined condition.

For example, whether or not the depth value is corrected may be determined based on the position, size, and movement of the overlay area. More specifically, the depth value in the depth map with respect to at least one of the overlay area and the background image when the overlay area is located at the center, the size of the overlay area is larger than a predetermined threshold value, or the overlay area does not move. Can be corrected.

In operation 340, the depth value of the depth map is corrected for at least one of the detected overlay region and the background image.

That is, in this embodiment, three types of depth value correction methods are performed as follows.

First, a method of correcting a depth value in a depth map only for a background image is performed.

In the first correction method, i) a method of correcting the objects of the background image to have the same depth value, or ii) a method of reducing the difference between the depth values of each of the objects of the background image may be used. In the case of ii), the difference between the depth values of the objects of the background image may be reduced by a predetermined ratio.

For example, when the first correction method is applied when the OSD is displayed on the screen, the background image may be displayed as a 2D image or close to the 2D image, so that the user may view the OSD area less tiredly.

The second method is to correct the depth value in the depth map only for the overlay area.

In the second correction method, i) a binocular parallax value of the overlay region is set to 0, ii) a depth value higher by a predetermined depth than the highest depth value among the depth values of the objects of the background image as the depth value of the overlay region. Setting method, iii) setting a depth value higher than a mean value of the depth values of objects of the background image as a depth value of the overlay area, iv) a depth value at the center of the overlay area or a depth at a boundary V) a method of setting a value as a depth value of the entire overlay area, v) a method of setting a depth value obtained by applying weights to the depth value of the background image and the depth value of the overlay area as the depth value of the overlay area, vi) any Setting a default value to a depth value of an overlay area, vii) a plurality of overlays when there are multiple overlay areas How to set the minimum, maximum or average value of the depth value of each area as the depth value of all overlay areas, viii) how to set the depth value input by the user as the depth value of the overlay area, ix) A method of correcting using a depth value of the background image may be used.

In this case, v) a method of setting the depth value obtained by applying weights to the depth value of the background image and the depth value of the overlay area as the depth value of the overlay area will be described with reference to Equation (1). For example, if the weights applied to the depth value of the background image and the depth value of the overlay area are 0.7 and 0.3, the average value (Dbaver) of the depth values of the objects of the background image is calculated, and then the depths of the objects of the background image are calculated. By multiplying the average value (Dbaver) of the values by 0.7, multiplying the depth value (Do) of the overlay area by 0.3, and summing both values, the corrected depth value (Dco) to be set as the new depth value of the overlay area is calculated. Can be.

[Equation 1]

Meanwhile, a method of correcting the depth value of the overlay area using the depth value of the background image around the overlay area of vii) will be described later with reference to FIG. 4.

On the other hand, if the depth value of the depth map is corrected only for the overlay region according to the second correction method, the depth value of the overlay region and the depth value of the background image are severely different, and distortion may occur at the boundary of the overlay region. Therefore, in another embodiment, a smoothing filter such as a median filter, a weighted average filter, or a maximum filter is applied to the overlay area in order to remove such distortion. Can be applied.

However, in one embodiment of the present invention, all kinds of filters capable of removing the difference between the depth value of the overlay area and the depth value of the background image may be used in addition to the smoothing filter. For example, in other embodiments a lowpass filter may be used instead of a smoothing filter.

Third, the depth value of the depth map is corrected for both the background image and the overlay area.

In another embodiment, it may be determined which correction method is performed from among the first to third correction methods described above according to the type of the overlay area.

For example, if the overlay area is a subtitle area, only the depth value of the overlay area is corrected. If the overlay area is an OSD area, only the depth value of both the overlay area and the background image is corrected, or only the depth value of the background image is corrected. You can correct it.

In operation 350, binocular images of left and right eyes are generated based on the depth map completed by the correction process in step 340, or both eyes images are generated based on the depth map of step 310 in which the depth value is not corrected.

In this case, the binocular images of the left and right eyes may be generated by referring to only pixel values in the background image without referring to pixel values in the overlay area.

FIG. 4 is a flowchart for describing an exemplary embodiment of correcting a depth value of an overlay region in FIG. 2.

Referring to FIG. 4, a method of using an average depth value of a surrounding background is disclosed as an example of correcting a depth value of an overlay area.

Referring to each operation process, first, at least one depth value is searched for a background image located at a boundary of the detected overlay area (410), and an average value of the at least one detected depth value is calculated (420). An average value may be uniformly assigned to the depth value of the overlay area (430).

In addition, it is also possible to search not only the average value but also the depth values of the surrounding area surrounding the overlay area and apply their median value, mode, and the like.

5A is a diagram illustrating an example in which an embodiment of the present invention is not applied when converting an overlay region into a 3D image. 5B is a diagram illustrating an example of applying an embodiment of the present invention when converting an overlay region into a 3D image.

Referring to FIG. 5A, first, a left eye image 510 and a right eye image 520 of an overlay area are shown. Comparing the positions of the subtitles represented in the left and right eye images with a1, a2, and a3 (where the solid line indicates the position of each letter in the left eye image 510, and the dotted line indicates the position of each character in the lower right eye image 520). It can be seen that the binocular disparity applied to each letter is expressed differently. This can be clearly seen by comparing the corresponding parts b1, b2, and b3 in FIG. 5B. When the subtitles are three-dimensionally displayed in the state as shown in FIG. 5A, it is difficult for the viewer to read the subtitles and adds fatigue during long time viewing.

However, referring back to FIG. 5B to which an embodiment of the present invention is applied, when comparing the positions of the captions represented in the left eye image 510 and the right eye image 520 with respect to the overlay area, b1, b2, and b3, respectively, It can be seen that the binocular parallax applied to the letters of U are uniformly represented (this can be clearly seen in comparison with corresponding parts a1, a2, and a3 in FIG. 5A). When the subtitles are three-dimensionally displayed in the state as shown in FIG. 5B, the distortion does not occur in the overlay area, and the viewer can read the subtitles more comfortably.

6 is a functional block diagram illustrating a 3D image converting apparatus of an overlay area according to another embodiment of the present invention.

Referring to FIG. 6, an embodiment of the 3D image conversion apparatus 600 of an overlay area includes a depth map generator 610 and an input monocular for generating a depth map indicating a perspective from an input monocular image. An overlay region detector 620 for detecting an overlay region included in an image, a depth value corrector 630 for correcting a depth value in a depth map, and a corrected depth map for at least one of the detected overlay region and a background image And a binocular image generator 640 for generating a binocular image.

Meanwhile, the depth value corrector 630 may include a searcher for searching for at least one or more depth values with respect to a background image located at a boundary of the detected overlay area, and an average value calculator for calculating an average value of the at least one or more searched depth values. It may further include.

On the other hand, the 3D image conversion method of the overlay area according to an embodiment of the present invention can be written in a program that can be executed in a computer, in a general-purpose digital computer to operate the program using a computer-readable recording medium Can be implemented.

In addition, as described above, the structure of the data used in the present invention may be recorded on a computer-readable recording medium through various means.

The computer-readable recording medium may include a storage medium such as a magnetic storage medium (eg, a ROM, a floppy disk, a hard disk, etc.) and an optical reading medium (eg, a CD-ROM, a DVD, etc.).

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

Claims (33)

In the 3D image conversion method of the overlay area (Overlaying area),
Generating a depth map indicating a perspective from the input monocular image;
Detecting an overlay region included in the input monocular image;
Correcting a depth value in the depth map with respect to at least one of the detected overlay area and the background image;
And generating a binocular image based on the corrected depth map. The method of claim 1,
The correcting of the 3D image converting method may include correcting a depth value close to a depth value allocated to a peripheral area of the detected overlay area. The method of claim 2,
The correcting step,
Searching for at least one depth value with respect to a background image located at a boundary of the detected overlay area;
Calculating an average value of the retrieved at least one depth value;
And uniformly allocating the calculated average value as a depth value of the overlay area. The method of claim 2,
Detecting the overlay area,
3. The method of claim 1, wherein an overlay region is extracted by analyzing edges and color components in the input monocular image. The method of claim 2,
The step of generating the binocular image,
And a binocular image of left and right eyes is generated by calculating binocular disparity from the completed depth map by reflecting the correction value for the overlay area. The method of claim 1,
Detecting the overlay area is performed based on position information of the overlay area;
3. The method of claim 3, wherein the position information of the overlay region is coordinates indicating a position of the overlay region or bitmap information indicating a position of the overlay region. The method of claim 1,
The correcting step may further include determining whether to perform the correction based on at least one of the position, size, and movement of the overlay area.
And the correcting is performed only when it is determined to perform the correction as a result of the determination. The method of claim 1,
The correcting step,
3D image conversion method, characterized in that to correct all the objects of the background image have the same depth value. The method of claim 1,
The correcting step,
3D image conversion method characterized in that the correction so that the difference between the depth value of each of the objects of the background image is reduced. The method of claim 1,
The correcting step,
A depth value higher than a mean value of depth values of the objects of the background image or a depth higher than a mean value of the depth values of the objects of the background image as a depth value of the overlay area; Transformation method. The method of claim 1,
The correcting step
3. The method of claim 3, wherein the depth value at the center of the overlay area or the depth value at the boundary of the overlay area is set as the depth value of the entire overlay area. The method of claim 11,
And after the correction is performed, performing filtering on the overlay area to remove a difference between the depth value of the overlay area and the depth value of the background image. The method of claim 1,
The correcting step
And a depth value obtained by applying weights to the depth value of the background image and the depth value of the overlay area, respectively. The method of claim 1
The correcting step
3. The method of claim 3, wherein when there are a plurality of overlay regions, one of minimum, maximum, and average values of depth values of each of the plurality of overlay regions is set as a depth value of each of the plurality of overlay regions. The method of claim 1,
The correcting step
And a depth value input from a user as a depth value of the overlay area. The method of claim 1,
The step of generating the binocular image
3. The method of claim 3, wherein the pixel values in the overlay area are referred to without referring to pixel values in the background image. In the 3D image conversion apparatus of the overlay area,
A depth map generator for generating a depth map displaying a perspective from the input monocular image;
An overlay region detector for detecting an overlay region included in the input monocular image;
A depth value corrector for correcting a depth value in the depth map with respect to at least one of the detected overlay area and the background image;
And a binocular image generator configured to generate a binocular image based on the corrected depth map. The method of claim 17,
And the depth value correcting unit corrects the depth value of the overlay area to a value close to the depth value allocated to the detected peripheral area of the overlay area. The method of claim 18, wherein the depth value correction unit,
A searcher for searching for at least one depth value with respect to a background image located at a boundary of the detected overlay area;
Further comprising an average value calculator for calculating an average value of the at least one searched depth value,
And uniformly assigning the calculated average value to a depth value of the overlay area. The method of claim 18,
The overlay area detector,
And an overlay region is extracted by analyzing edges and color components from the input monocular image. The method of claim 18,
The binocular image generation unit,
And a binocular image of left and right eyes is generated by calculating binocular disparity from the completed depth map by reflecting the correction value for the overlay area. The method of claim 17,
The overlay area detector
Detecting the overlay region based on position information of the overlay region;
And the position information of the overlay region is coordinates indicating the position of the overlay region or bitmap information indicating the position of the overlay region. The method of claim 17,
The depth value corrector determines whether to perform the correction based on at least one of the position, size, and movement of the overlay area, and performs the correction only when it is determined that the correction is performed. 3D video conversion device characterized in that. The method of claim 17,
The depth value correction unit,
3D image conversion apparatus, characterized in that for correcting so that all the objects of the background image have the same depth value. The method of claim 17,
The depth value correction unit,
And correcting the difference between depth values of the objects of the background image to be reduced. The method of claim 17,
The depth value correction unit,
A depth value higher than a mean value of depth values of the objects of the background image or a depth higher than a mean value of the depth values of the objects of the background image as a depth value of the overlay area; Converter. The method of claim 17,
The depth value correction unit
And a depth value at a center of the overlay area or a depth value at a boundary of the overlay area as a depth value of the entire overlay area. The method of claim 27,
The depth value correction unit
And after performing the correction, further performing filtering on the overlay region to remove a difference between the depth value of the overlay region and the depth value of the background image. The method of claim 17,
The depth value correction unit
And performing the correction using a depth value obtained by applying weights to the depth value of the background image and the depth value of the overlay region, respectively. The method of claim 17
The depth value correction unit
3. The apparatus of claim 3, wherein when the overlay regions are plural, one of the minimum, maximum, and average values of the depth values of each of the plurality of overlay regions is set as the depth value of each of the plurality of overlay regions. The method of claim 17,
The depth value correction unit
And a depth value input from a user as a depth value of the overlay area. The method of claim 17,
The binocular image generation unit
And the binocular image is generated by referring to only pixel values in the background image without referring to pixel values in the overlay region. A computer-readable recording medium having recorded thereon a program for implementing the 3D image conversion method according to any one of claims 1 to 16.

Images