JP2013156955A - Image processing device, image processing method, and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate a high-quality 3D image while reducing a producer's time and effort when generating a 3D image from a 2D image.SOLUTION: A depth map generation unit 10 generates a depth map of an input image on the basis of the input image and a depth model. A depth map processing unit 20 processes the depth map generated by the depth map generation unit 10 for each of a plurality of areas within a screen specified by a plurality of externally set mask patterns. A 3D image generation unit 30 generates an image at another viewpoint on the basis of the input image and the depth map processed by the depth map processing unit 20. A display control unit 50 displays, on a display in a graph, data representing the state of depths in the plurality of specified areas in the depth map respectively processed by the depth map processing unit 20.

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program for executing processing for converting a 2D image into a 3D image for stereoscopic viewing.

  In recent years, 3D video contents such as 3D movies and 3D broadcasts have become widespread. In order to make an observer stereoscopically view, a right eye image and a left eye image having parallax are required. When displaying a 3D image, the right eye image and the left eye image are displayed in a time-sharing manner, and the right eye image and the left eye image are separated by image separation glasses such as shutter glasses and polarized glasses. Thus, the observer can observe the right eye image only with the right eye and the left eye image only with the left eye and can perform stereoscopic viewing. Note that if the right eye image and the left eye image are divided not by time division but by space division, glasses are not necessary, but the resolution is lowered. Both the glasses method and the glassesless method are common in that a right eye image and a left eye image are required.

  There are two main methods for producing 3D video. A method for simultaneously capturing a right eye image and a left eye image using two cameras, and a later editing of a 2D image captured by one camera. There is a method for generating a parallax image. The present invention relates to the latter method, and to 2D3D conversion technology.

  FIG. 1 is a diagram for explaining a basic processing process of 2D3D conversion. First, a depth map (also referred to as depth information) is generated from the 2D input image (step S10). Then, a 3D image is generated using the 2D input image and the depth map (step S30). In FIG. 1, a 2D input image is a right eye image of a 3D output image, and an image obtained by pixel shifting the 2D input image using a depth map is a left eye image of the 3D output image. Hereinafter, a combination of a right eye image and a left eye image having a predetermined parallax is referred to as a 3D image or a parallax image.

  Conventionally, various methods for estimating a scene structure expressed by a 2D image and generating a depth map have been proposed. The applicant calculates the statistic of pixel values in a predetermined area on the screen to estimate the scene structure, determines the composition ratio of multiple basic depth models accordingly, and generates a depth map according to the composition ratio We propose a method to do this. If this method is used, a 3D image with relatively little discomfort can be easily and quickly generated from a 2D image.

JP 2011-48586 A

  High quality 3D images are required in the field of video production such as 3D movies. There is a limit to the quality of 3D images that can be generated by the above-described method. Enormous human work is required to produce high-quality 3D images. For example, detailed operations such as setting and adjusting the bulge of each of a large number of objects on the screen, the positional relationship between the objects, and adding gradation are required. The effort and cost for performing such operations on all frames is enormous.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a technology capable of generating a high-quality 3D image while reducing the effort of the producer when generating a 3D image from a 2D image. It is in.

  In order to solve the above problems, an image processing apparatus according to an aspect of the present invention includes a depth map generation unit (10) that generates a depth map of the input image based on the input image and a depth model, and the depth map. A depth map processing unit (20) that processes the depth map generated by the generation unit (10) for each of a plurality of regions in a screen specified by a plurality of mask patterns set externally; the input image; and Based on the depth map processed by the depth map processing unit (20), the image generation unit (30) for generating an image of another viewpoint, and the depth map processed by the depth map processing unit (20), respectively. And a display control unit (50) for displaying data representing the state of depth in a plurality of designated areas in a graph on a display.

  Another aspect of the present invention is an image processing method. The method includes a step of generating a depth map of the input image based on the input image and a depth model, and the generated depth map is converted into a plurality of images in a screen specified by a plurality of externally set mask patterns. A step of processing for each region, a step of displaying on the display data representing the state of depth in each of the plurality of specified regions in the depth map, the input image, and the processed depth map Originally, generating an image of a different viewpoint.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as an aspect of the present invention.

  According to the present invention, when generating a 3D image from a 2D image, it is possible to generate a high-quality 3D image while reducing the labor of the producer.

It is a figure for demonstrating the basic processing process of 2D3D conversion. It is a figure which shows the structure of the image editing system which concerns on embodiment of this invention. It is a figure which shows the structural example of the depth map production | generation part which concerns on embodiment of this invention. It is a figure for demonstrating the whole processing process of the image editing system which concerns on embodiment of this invention. It is a figure for demonstrating the gain adjustment process of an input depth map. It is a figure for demonstrating the synthetic | combination process of a layer depth map. It is a figure for demonstrating the gain adjustment process of the input depth map which does not use a mask. It is a figure for demonstrating the 3D image generation process by a 3D image generation part. It is a figure which shows the example which adjusts a gain centering on arbitrary levels. It is a figure which shows the example which adjusts a gain centering on the average value of a depth value. It is a figure which shows the example which multiplies the gain of a different value up and down centering on the average value of a depth value. It is a figure which shows the example which adjusts a gain centering on the value set externally. It is a figure for demonstrating the offset adjustment process of an input depth map. It is a figure for demonstrating the offset adjustment process of the input depth map which does not use a mask. FIGS. 15A to 15C are diagrams for explaining the dynamic range adjustment of the entire depth map. It is a figure for demonstrating the gradation addition process of an input depth map. It is a figure for demonstrating the gradation addition process of the input depth map which does not use a mask. It is a figure which shows an example of the graph which displays the depth state for every layer. It is a figure which shows the example which displays the depth state of a layer with a histogram.

  FIG. 2 is a diagram showing a configuration of an image editing system 500 according to the embodiment of the present invention. An image editing system 500 according to the embodiment includes an image processing device 100 and a console terminal device 200.

  The console terminal device 200 is a terminal device used by an image producer (hereinafter referred to as a user) to produce and edit an image. The console terminal device 200 includes an operation unit 60 and a display unit 70. The operation unit 60 is an input device such as a keyboard and a mouse, and the display unit 70 is an output device such as a display. Note that a touch panel display which is a device in which input and output are integrated may be used. The console terminal device 200 may include a user interface that uses a printed material such as a printer or a scanner as a medium. The operation unit 60 receives a user operation, generates a signal resulting from the user operation, and outputs the signal to the image processing apparatus 100. The display unit 70 displays an image generated by the image processing apparatus 100.

  The image processing apparatus 100 includes a depth map generation unit 10, a depth map processing unit 20, a 3D image generation unit 30, an operation reception unit 40, and a display control unit 50. These configurations can be realized by an arbitrary processor, memory, or other LSI in terms of hardware, and can be realized by a program loaded into the memory in terms of software, but here by their cooperation. Draw functional blocks. Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof. For example, all the functions of the depth map generation unit 10, the depth map processing unit 20, and the 3D image generation unit 30 may be realized by software, and the functions of the depth map generation unit 10 and the 3D image generation unit 30 are dedicated logic. The depth map processing unit 20 may be configured by software and configured by a circuit.

  The depth map generation unit 10 generates a depth map of the 2D image based on the input 2D image and the depth model. The depth map is a grayscale image in which depth values (also referred to as depth values) are represented by luminance values. The depth map generation unit 10 estimates a scene structure and generates a depth map using a depth model suitable for the scene structure. In the present embodiment, the depth map generation unit 10 combines a plurality of basic depth models and uses them for depth map generation. At that time, the composition ratio of a plurality of basic depth models is changed according to the scene structure of the 2D image.

  FIG. 3 is a diagram illustrating a configuration example of the depth map generation unit 10 according to the embodiment of the present invention. The depth map generation unit 10 includes an upper screen high-frequency component evaluation unit 11, a lower screen high-frequency component evaluation unit 12, a composition ratio determination unit 13, a first basic depth model frame memory 14, and a second basic depth model frame memory 15. , A third basic depth model frame memory 16, a synthesis unit 17, and an addition unit 18.

  The screen upper high frequency component evaluation unit 11 calculates the ratio of pixels having a high frequency component in the upper screen of the 2D image to be processed. The ratio is used as the high frequency component evaluation value at the top of the screen. The ratio of the upper part of the screen to the entire screen is preferably set to approximately 20%. The lower screen high frequency component evaluation unit 12 calculates the ratio of pixels having a high frequency component in the lower screen of the 2D image. The ratio is used as the high frequency component evaluation value at the bottom of the screen. The ratio of the lower part of the screen to the entire screen is preferably set to approximately 20%.

  The first basic depth model frame memory 14 holds the first basic depth model, the second basic depth model frame memory 15 holds the second basic depth model, and the third basic depth model frame memory 16 stores the third basic depth model. Holds the depth model. The first basic depth model is a model in which the upper part of the screen and the lower part of the screen are concave spherical surfaces. The second basic depth model is a model in which the upper part of the screen is a cylindrical surface having an axis in the vertical direction and the lower part of the screen is a concave spherical surface. The third basic depth model is a model in which the upper part of the screen is a plane and the lower part of the screen is a cylindrical surface having an axis in the horizontal direction.

  The composition ratio determination unit 13 is a first basic depth model based on the high-frequency component evaluation values of the upper screen portion and the lower screen portion calculated by the upper screen region high-frequency component evaluation unit 11 and the lower screen region high-frequency component evaluation unit 12, respectively. The combination ratios k1, k2, and k3 (where k1 + k2 + k3 = 1) of the second basic depth model and the third basic depth model are determined. The synthesizing unit 17 multiplies the synthesis ratios k1, k2, and k3 by the first basic depth model, the second basic depth model, and the third basic depth model, and adds the multiplication results. This calculation result becomes a synthetic basic depth model.

  For example, when the high frequency component evaluation value at the upper part of the screen is small, the composition ratio determining unit 13 recognizes that the scene has an empty or flat wall at the upper part of the screen, and the second basic depth model with a deeper depth at the upper part of the screen. Increase the ratio. If the high-frequency component evaluation value at the bottom of the screen is small, it is recognized as a scene in which the flat ground or water surface continuously spreads at the bottom of the screen, and the ratio of the third basic depth model is increased. In the third basic depth model, the upper part of the screen is approximated as a distant view, and the depth is reduced toward the bottom of the lower part of the screen.

  The adding unit 18 generates a depth map by superimposing the red component (R) signal of the 2D image on the combined basic depth model generated by the combining unit 17. The reason for using the R signal is based on an empirical rule that there is a high probability of matching with the unevenness of the subject in an environment in which the magnitude of the R signal is close to direct light and the brightness of the texture is not significantly different. . Further, red and warm colors are forward colors in chromatics, and the depth is recognized in front of the cold color system, and the stereoscopic effect is emphasized.

  Returning to FIG. The depth map processing unit 20 processes the depth map generated by the depth map generation unit 10. In the present embodiment, the depth map processing unit 20 generates the depth map generated by the depth map generation unit 10 for each of a plurality of areas in the screen specified by a plurality of mask patterns (hereinafter simply referred to as masks) set externally. Process maps individually or independently. For example, processing such as gain adjustment, offset adjustment and gradation processing is performed. Details of the processing by the depth map processing unit 20 will be described later.

  The 3D image generation unit 30 generates a 2D image of another viewpoint based on the above-described 2D image and the depth map processed by the depth map processing unit 20. The 3D image generation unit 30 outputs the 2D image of the original viewpoint and the 2D image of the different viewpoint as the right eye image and the left eye image.

  Hereinafter, a specific example of generating a 2D image of another viewpoint having a parallax with the 2D image using the 2D image of the original viewpoint and the depth map will be described. In this specific example, a 2D image of another viewpoint that is moved to the left is generated based on the viewpoint when the 2D image of the original viewpoint is displayed on the screen. In that case, when displaying the texture as a foreground for the observer, move the texture of the 2D image of the original viewpoint to the left side toward the screen by a predetermined amount, and toward the screen when displaying the texture as a distant view for the observer. Move the texture to the right by a predetermined amount.

  Let Yd be the luminance value of each pixel in the depth map, m be the convergence value representing the pop-out feeling, and n be the depth value representing the stereoscopic effect. The 3D image generation unit 30 sequentially shifts the texture of the 2D image at the original viewpoint corresponding to the luminance value Yd to the left by (Yd−m) / n pixels in order from the smallest luminance value Yd. If the value of (Yd−m) / n is negative, shift to the right by (m−Yd) / n pixels. For the observer, a texture with a small luminance value Yd of the depth map appears on the back side of the screen, and a texture with a large luminance value Yd appears on the front side of the screen. The luminance value Yd, the convergence value m, and the depth value n are values in the range of 0 to 255. For example, the convergence value m = 200 and the depth value n = 20 are set.

  A more detailed description of the depth map generation by the depth map generation unit 10 and the 3D image generation by the 3D image generation unit 30 will be described in Japanese Patent Application Laid-Open Nos. 2005-151534 and 2009-44722 previously filed by the present applicant. It is disclosed in the gazette.

  The operation reception unit 40 receives a signal input from the operation unit 60 of the console terminal device 200. The display control unit 50 controls the display unit 70 of the console terminal device 200. In this specification, in order to pay attention to the processing of the depth map processing unit 20, in FIG. 2, only the line between the operation receiving unit 40 and the display control unit 50 and the depth map processing unit 20 is connected. The operation receiving unit 40 and the display control unit 50 are not limited to use for the processing of the depth map processing unit 20.

  FIG. 4 is a diagram for explaining the overall processing process of the image editing system 500 according to the embodiment of the present invention. In general, a 2D image includes a plurality of objects. The 2D input image of FIG. 4 includes three objects. Specifically, it includes people, trees, and background objects. First, the depth map generation unit 10 generates a depth map from the 2D input image (step S10). In the depth map, the closer to white, the higher the luminance and the closer to the observer, and the closer to black, the lower the luminance and the farther the observer. When a 3D image is generated, the pop-out amount increases as the area is closer to white in the depth map, and the retraction amount increases as the area is closer to black.

  In the present embodiment, in order to individually adjust the sense of depth for a plurality of objects in an image, effect adjustment is performed independently for each object area in the depth map. Specifically, each object area is specified in the depth map using a plurality of masks representing each object area in the image. Then, effect adjustment is performed individually for each identified object region, and a plurality of depth maps adjusted for effect are obtained. Then, one depth map is generated by combining the plurality of depth maps. This depth map is used when generating a 2D image of another viewpoint from a 2D image of the original viewpoint.

  The depth map generator 10 automatically generates a depth map of the 2D input image (S10). The generated depth map is input to the depth map processing unit 20. The depth map processing unit 20 also receives a plurality of masks respectively representing a plurality of object regions in the 2D input image. These masks are generated based on the outline of the object area traced by the user. For example, the display control unit 50 displays a 2D input image on the display unit 70, and the user traces the outline of the region to be an object region in the 2D input image using the operation unit 60. The operation receiving unit 40 generates contour information of each object region based on a signal from the operation unit 60 and outputs the contour information to the depth map processing unit 20 as a mask. Note that the mask may be read by the image processing apparatus 100 by reading the outline drawn on the printed matter by the scanner.

  In FIG. 4, the effective area of each mask is drawn in white and the ineffective area is drawn in black. The mask of the person is a pattern in which only the person area is valid and the other areas are invalid. The tree mask is a pattern in which only the tree area is valid and the other areas are invalid. The background mask is a pattern in which only the background is valid and other areas are invalid.

  There is no restriction on the number of masks per screen, and the user can arbitrarily set it. The object area can be set to an area determined by the user to be one object area. For example, as shown in FIG. 4, one object area may be set for one person, or an object area may be set for each part of the person and for each part. In particular, when generating a high-quality 3D image, a large number of object regions can be set for one person, and the position in the thickness and depth direction can be adjusted for each part and for each part. is there.

  The depth map processing unit 20 processes the depth map (hereinafter referred to as input depth map) input from the depth map generation unit 10 using a plurality of masks input via the user interface (S20). The depth map processing unit 20 processes the depth map individually for each area specified by each mask. Hereinafter, this depth map processing for each region is referred to as depth map layer processing. A depth map that has been subjected to layer processing is called a layer depth map. In this specification, a layer is used as a concept indicating a unit of processing for an effective area of a mask.

  In FIG. 4, as an example, the depth map processing unit 20 performs layer processing by specifying a person region from the input depth map using a person mask (layer 1 mask) (S <b> 21 a). Similarly, from the input depth map, a tree region is identified using a tree mask (layer 2 mask) and layer processing is performed (S21b). Similarly, a background region is identified from the input depth map using a background mask (layer 3 mask) and layer processing is performed (S21c).

  The depth map processing unit 20 synthesizes the depth maps of the object areas of the layer depth maps of layers 1 to 3 (S22). This combined depth map is referred to as a combined depth map. The 3D image generation unit 30 shifts the pixels of the 2D input image using the composite depth map, and generates an image having a parallax with respect to the 2D input image (S30). The 3D image generation unit 30 outputs the 2D input image as the right eye image (R) of the 3D output image and the generated image as the left eye image (L).

(Gain adjustment)
First, an example of gain adjustment as layer processing by the depth map processing unit 20 will be described. Gain adjustment is a process of adjusting the thickness of the object in the depth direction. Increasing the gain makes the object thicker and lowering the gain makes the object thinner.

  FIG. 5 is a diagram for explaining the gain adjustment process of the input depth map. The depth map processing unit 20 multiplies the input depth map, which is a depth map before processing, by a gain only in the effective area of the person's mask, and increases the amplitude of the depth value only in the portion of the person in the input depth map. (S21a). In FIG. 5, the amplitude of the person portion of the layer depth map, which is a depth map after processing, is large (see symbol a).

  FIG. 6 is a diagram for explaining a layer depth map synthesis process. The depth map processing unit 20 cuts out only the effective area of the layer 1 mask (person's mask) from the layer 1 layer depth map (person's depth map). Similarly, only the effective area of the layer 2 mask (tree mask) is cut out from the layer 2 layer depth map (tree depth map). Similarly, only the effective area of the layer 3 mask (background mask) is cut out from the layer 3 layer depth map (background depth map). The depth map processing unit 20 generates a combined depth map by combining the three extracted depth maps.

  FIG. 7 is a diagram for explaining an input depth map gain adjustment process without using a mask. If the layer depth map is synthesized by using only the effective area of the mask of each layer depth map, the invalid area of the mask of each layer depth map is not reflected in the synthesized depth map. Therefore, the depth map processing unit 20 multiplies the entire input depth map by the gain to increase the amplitude of the depth value of the entire input depth map (S21a). In FIG. 7, the entire amplitude of the layer depth map is large (see symbol b).

  FIG. 8 is a diagram for explaining a 3D image generation process by the 3D image generation unit 30. The 3D image generation unit 30 shifts the pixels of the 2D input image based on the combined depth map, and generates an image having a parallax with respect to the 2D input image (S30). FIG. 8 shows an example of shifting the pixel of the human part in the 2D input image to the left. The 3D image generation unit 30 increases the depth value by adding an offset value to the depth value of the person portion of the combined depth map. Details of the offset adjustment will be described later. When the depth value of the person portion increases, the pop-out amount of the person portion of the 3D image increases.

  If only the person portion is pixel-shifted without shifting the background portion around the person portion, a missing pixel portion in which no pixel exists is generated (see reference symbol c of the pixel-shifted image before correction). The 3D image generation unit 30 corrects the missing pixel portion by interpolating the missing pixel portion with a pixel generated from the surrounding pixels. There are various methods for pixel interpolation. For example, interpolation is performed using pixels at the boundary of the person portion (see reference numeral d of the pixel shifted image after correction).

  The purpose of multiplying the depth value in the depth map to expand or compress the amplitude of the depth value is to adjust the amplitude of the sense of depth in the generated 3D image. Therefore, the center level for multiplying the depth value by gain needs to be set appropriately.

  FIG. 9 is a diagram illustrating an example of gain adjustment centering on an arbitrary level. In FIG. 9, an arbitrary level is set to a value lower than the average value of the depth values in the object area of the depth map. As described above, the object area of the depth map is specified by the effective area of the mask. When the gain is adjusted around the arbitrary level, the average value of the depth values fluctuates and an unintended offset occurs in the depth map. This offset changes not only the amplitude of the sense of depth of the generated 3D image but also the center of gravity of the depth. This change in the center of gravity is an unintended adjustment for the user. In FIG. 9, after the gain adjustment, the average value of the depth value becomes high, and the center of gravity of the depth shifts to the near side.

  FIG. 10 is a diagram illustrating an example in which gain adjustment is performed centering on an average value of depth values. The depth map processing unit 20 sets the center level for gain adjustment to the average value of the depth values in the object area of the depth map. The gain value used for gain adjustment is set from the outside. The user inputs a desired gain value using the operation unit 60, and the operation reception unit 40 receives the gain value input from the operation unit 60 and sets it in the depth map processing unit 20.

  As shown in FIG. 10, when the gain adjustment is performed with the average value of the depth values as the center level, the average value of the depth values does not vary before and after the adjustment, so that an unintended offset does not occur in the depth map. Therefore, only the amplitude of the sense of depth of the generated 3D image changes, and the center of gravity of the depth does not change.

  The average depth value can be obtained by the following procedure. First, the sum of depth values in the object area of the depth map is calculated. Next, the area in the object region, that is, the number of pixels is counted. Finally, the sum of the depth values in the object area is divided by the number of pixels in the object area. Thereby, the average value of the depth values in the object area is calculated.

  When the depth value in the depth map is multiplied by a gain to expand or compress the amplitude of the depth value, it may be desired to adjust the gain independently on the protruding side and the retracting side.

  FIG. 11 is a diagram illustrating an example in which gains of different values are multiplied up and down around the average value of depth values. The depth map processing unit 20 adjusts the gain independently above and below the center level. More specifically, with the average value of the depth values in the object area of the depth map as the central level, the first gain value that multiplies the depth value that is larger than the average value and the depth value that is smaller than the average value are multiplied. The second gain value to be set is set independently. In FIG. 11, the first gain value is set larger than the second gain value (see symbols e and f).

  When the depth value of the target pixel is larger than the average value, the depth map processing unit 20 multiplies the depth value by the first gain value. When the depth value of the target pixel is smaller than the average value, the depth map processing unit 20 sets the depth value to the depth value. Multiply the second gain value. Thereby, the gain can be adjusted independently on the protruding side and the depth side. In FIG. 11, the thickness on the protruding side is increased.

  FIGS. 10 and 11 show examples in which gain adjustment is performed using the average value of depth values obtained by calculation as the center level. In this regard, the center level for gain adjustment may be given from the outside.

  FIG. 12 is a diagram illustrating an example of gain adjustment centering on an externally set value. The depth map processing unit 20 sets the center level for gain adjustment to a designated reference level input from the outside. The user inputs the designated reference level using the operation unit 60, and the operation receiving unit 40 receives the designated reference level input from the operation unit 60 and outputs it to the depth map processing unit 20. As a result, the user can arbitrarily designate the center level for gain adjustment.

  10 to 12 can be selected for each object area of the depth map. In other words, a method for setting the center level for gain adjustment and a method for changing the gain value above and below the center level can be selected for each object area of the depth map. Although FIGS. 10 to 12 illustrate examples in which these methods are applied to the object region of the depth map, these methods may be applied to the synthesized depth map.

(Offset adjustment)
Next, an example in which offset adjustment is performed as layer processing by the depth map processing unit 20 will be described. Offset adjustment is a process of adjusting the position of the object in the depth direction. When a positive offset value is added, the object moves in the pop-out direction, and when a negative offset value is added, the object moves in the retracting direction.

  FIG. 13 is a diagram for explaining the offset adjustment process of the input depth map. The depth map processing unit 20 adds an offset only to the effective area of the tree mask to the input depth map, which is the depth map before processing, and increases the level of the depth value only for the portion of the tree in the input depth map. (S21b). In FIG. 13, the layer depth map, which is the depth map after processing, has a high level of the tree portion (see symbol g).

  FIG. 14 is a diagram for explaining an offset adjustment process of an input depth map that does not use a mask. If the layer depth map is synthesized by using only the effective area of the mask of each layer depth map, the invalid area of the mask of each layer depth map is not reflected in the synthesized depth map. Therefore, the depth map processing unit 20 adds an offset to the entire input depth map to increase the level of the depth value of the entire input depth map (S21b). In FIG. 14, the overall level of the layer depth map is high (see symbol h).

  The offset value used for offset adjustment is set from the outside. The user inputs a desired offset value using the operation unit 60, and the operation reception unit 40 receives the offset value input from the operation unit 60 and sets it in the depth map processing unit 20. In the above description, the offset value is added to the object area of the depth map. However, the offset value may be added to the composite depth map.

  If the gain of the composite depth map is adjusted instead of the offset area of the depth map, gain adjustment as intended may not be possible. Hereinafter, it is assumed that the dynamic range of the depth value of the depth map is enlarged or reduced to increase or decrease the depth feeling of the entire screen. In the following description, a case where the feeling of depth is increased will be described, but the same applies to the case where the feeling of depth is weakened.

  As a method for enhancing the sense of depth of the entire screen, a method of applying a uniform gain to the synthesized depth map can be considered. However, in this method, the dynamic range of the depth value is expanded, and at the same time, the amplitude of the unevenness of each object is also expanded. The latter enlargement may also be suitable for the purpose of adjustment, but it may not be.

  Hereinafter, an adjustment method in which the dynamic range of the depth value is expanded but the amplitude of the unevenness of each object is not changed will be described. The depth map processing unit 20 is common for adjusting the dynamic range of the entire depth map to each difference between the average value of the depth value of the entire depth map and the average value of the depth value in each object region in the depth map. Multiply by a factor of. Each calculation result becomes an offset value of the depth value in each object area. The depth map processing unit 20 adds each corresponding offset value to the depth value in each object region.

  Hereinafter, a specific example will be described. It is assumed that the depth values of the combined depth map are distributed in −X to + X, and this depth value distribution −X to + X is set as a dynamic range of the depth values. Consider an example of expanding the dynamic range of depth values. For example, the dynamic range of the depth value is increased by a factor of 1.5, and the depth value distribution is changed from -X to + X to -1.5X to + 1.5X. The following processing is executed in order to expand the dynamic range of the depth value without changing the amplitude of the unevenness of each object.

  First, the depth map processing unit 20 calculates the minimum value, the maximum value, and the average value of the depth values in the entire depth map. The difference between the minimum value and the maximum value is the dynamic range of the depth value. Next, the depth map processing unit 20 calculates an average value of the depth values in each object area of the depth map. Next, the depth map processing unit 20 subtracts the average value of the depth values in the entire depth map from the average value of the depth values in each object region. This subtraction result is called an average difference of layer depths.

  In the following, let us consider a case where the dynamic range of the depth value is to be expanded a times. The depth map processing unit 20 multiplies the average difference of each layer depth by a. Next, the average difference of each layer depth is subtracted from the average difference of each layer depth multiplied by a. This subtraction result is referred to as an average difference offset value for each layer depth. Finally, the depth map processing unit 20 adds the offset value of the average difference of each layer depth to the depth value in each object area of the depth map.

  Thereby, only the offset value of the depth value changes in each layer depth map, and the amplitude value of the depth value does not change. Therefore, the dynamic range of the depth value can be expanded without changing the amplitude of the unevenness of each object.

  FIGS. 15A to 15C are diagrams for explaining the dynamic range adjustment of the entire depth map. The example in which the number of layers in the 2D input image is 3 has been described so far, but FIGS. 15A to 15C show examples in which the number of layers is 5 (layer 0 to layer 4). When viewed from the observer, the layer 0 object appears most retracted, and the layer 4 object appears most popped out.

  FIG. 15A shows an example of the depth of each layer before dynamic range adjustment. An arrow (see symbol 1) in each layer is defined by a minimum value (see symbol j) and a maximum value (see symbol k) in the object area of each layer. That is, the arrow (see symbol 1) indicates the dynamic range of each layer. A line connecting the arrows is a line connecting the average values of the depth values (see symbol i) in the object areas of the layers.

  FIG. 15 (b) shows the depth of each layer after applying twice the gain adjustment uniformly to the depth of each layer of FIG. 15 (a). The dynamic range of the entire depth map is doubled, and at the same time, the dynamic range of the object area of each layer, that is, the amplitude of the unevenness of each layer is doubled.

  FIG.15 (c) shows the depth of each layer after applying the offset adjustment mentioned above to the depth of each layer of Fig.15 (a). Although the dynamic range of the entire depth map is doubled, the dynamic range of the object area of each layer, that is, the amplitude of the unevenness of each layer does not change. For example, when there are a plurality of persons as objects on the screen, the depth feeling can be enhanced by adjusting the distance between the persons without changing the thickness of each person.

(Gradation added)
Next, an example in which gradation is added as a layer process by the depth map processing unit 20 will be described. The addition of gradation is a process for inclining the depth of each object.

  FIG. 16 is a diagram for explaining the gradation addition process of the input depth map. The depth map processing unit 20 adds a gradation to only the effective area of the background mask (only the ground in FIG. 16) with respect to the input depth map which is the depth map before processing, and only the background portion is inclined to the depth value. (S21c). In FIG. 16, the layer depth map which is the depth map after processing has an inclination in the depth of the ground (see symbol m).

  More specific description will be given below. A gradation pattern holding unit (not shown) holds at least one gradation pattern generated in advance. As shown in FIG. 16, the gradation pattern holding unit may hold not only a flat gradation pattern but also various shapes such as a spherical surface and a cylindrical shape. Even if the gradation pattern has the same shape, a plurality of gradation patterns having different inclination angles and / or inclination directions may be held.

  The user can use the operation unit 60 to display a plurality of gradation patterns held in the gradation pattern holding unit on the display unit 70 and select one of them. The operation receiving unit 40 receives an input from the operation unit 60, acquires a gradation pattern corresponding to the input, and outputs the gradation pattern to the depth map processing unit 20.

  The depth map processing unit 20 receives an input depth map from the depth map generation unit 10, a mask from the outside, and a gradation parameter from the gradation pattern holding unit. The depth map processing unit 20 adds a gradation pattern to the effective area of the mask in the input depth map.

  FIG. 17 is a diagram for explaining a gradation addition process of an input depth map that does not use a mask. If the layer depth map is synthesized by using only the effective area of the mask of each layer depth map, the invalid area of the mask of each layer depth map is not reflected in the synthesized depth map. Therefore, the depth map processing unit 20 adds gradation to the entire input depth map, and inclines the depth value of the entire input depth map (S21c). In FIG. 17, the layer depth map is inclined as a whole (see symbol n).

  In the above description, the example in which the gradation pattern to be added to the depth map is selected from typical gradation patterns generated in advance has been described. Hereinafter, an example in which the gradation pattern to be added to the depth map is specified by the control parameter will be described. The depth map processing unit 20 adds a gradation to the object area of the depth map based on the gradient parameter and the direction parameter of the gradation that can be independently adjusted. Furthermore, gradation can be added while adjusting the offset of the designated area of the depth map based on the offset parameter set externally.

  The user can input the gradation parameter, direction parameter, and offset parameter of the gradation using the operation unit 60. The operation reception unit 40 receives an input from the operation unit 60 and outputs the input to the depth map processing unit 20.

The process of adding gradation to the depth map can be realized by adding a gradation depth value corresponding to the angle and inclination of the gradation to each pixel of the depth map. Slope and Angle are used as control parameters for performing this processing. Slope is defined by a change value of the depth value per pixel. Hereinafter, in order to simplify the explanation, an inclination model in which the depth value changes in a proportional function will be considered. There are various other types of gradient models, such as the depth value changing exponentially.
Slope: Gradient slope [depth value / pixel]
Angle: Angle relative to the gradation image [degree]

The gradation is added with the center of the screen as a reference, and the coordinates of the center of the screen are (x, y) = (0, 0). Based on this, the coordinates of an arbitrary pixel on the image are (x, y) = (x_base, y_base). If the gradient slope values in the X-axis direction and Y-axis direction are slope_x and slope_y, slope_x and slope_y are expressed by the following formulas (1) and (2).
slope_x = slope * cosθ = slope * cos (2π * (Angle / 360)) ... Formula (1)
slope_y = slope * sinθ = slope * sin (2π * (Angle / 360)) ... Formula (2)

Assuming that the gradation depth values in the X-axis direction and Y-axis direction at a certain coordinate are grad_depth_x and grad_depth_y, grad_depth_x and grad_depth_y are expressed by the following equations (3) and (4).
grad_depth_x = Slope_x * x_base (3)
grad_depth_y = Slope_y * y_base (4)

In order to add gradation to the depth map, assuming that the gradation depth value added to the depth value of each pixel is grad_depth, grad_depth is expressed by the following equation (5).
grad_depth = grad_depth_x + grad_depth_y (5)

  The gradation can be added to the depth map by adding the gradation depth value obtained by the above equation (5) to each pixel of the depth map. The gradient slope can be freely set by changing the Slope value, and the angle of the gradation image by changing the Angle value.

  As described above, the method for adding gradation to the depth map using the equations (1) to (5) has been described. In this method, since the gradation depth value is added based on the center of the screen, the depth value of the depth map does not change at the center of the screen. Hereinafter, a method of adding a gradation to an arbitrary position by adding an offset at the same time as adding a gradation to the depth map will be described.

Offset is used as a parameter to control the offset value to be added.
Offset: Gradient offset

When a gradation depth value including an offset value to be added to each pixel when adding gradation to the depth map is grad_offset_depth, grad_offset_depth is expressed by the following equation (6).
grad_offset_depth = grad_depth_x + grad_depth_y + Offset Equation (6)

  Thereby, an arbitrary level of gradation can be added at an arbitrary position on the screen. In the above description, an example in which gradation is added to the object area of the depth map has been shown. However, gradation may be added to the combined depth map.

(User interface)
When performing the operation of generating a 3D image, it is necessary to give a correct parallax to the generated 3D image so that the object context of the generated 3D image is correct. For this purpose, it is necessary to correctly process a depth map that holds disparity information that is a reference for generating disparity. The user is working while displaying which depth map is displayed on the display and checking which part of the image has a certain amount of parallax.

  In this case, it is relatively easy to check the difference in the amount of parallax between adjacent portions or adjacent portions on the depth map. On the other hand, it is often difficult to confirm the difference in the amount of parallax for the distant portions. Although it is conceivable to display the parallax amount as a numerical value and read it, there is a concern that the efficiency of the confirmation work may be reduced as compared with the case where the parallax amount is confirmed on the image because the data amount of the image is very large.

  In addition, by observing only the combined depth map, it is not possible to determine which area of the depth map corresponds to which layer. It is necessary to observe the mask of each layer simultaneously. Since there are a plurality of layer masks, it is necessary to observe a plurality of layer masks and a combined depth map at the same time. However, even by this observation, it is difficult to clearly confirm the difference in the amount of parallax between the distant parts. Thus, in practice, it is difficult to clearly grasp which area of the composite depth map corresponds to which layer.

  The present inventor has developed a user interface suitable for grasping depth values of all layers and grasping differences from other layers. This user interface can be used instead of or in addition to a user interface that displays a composite depth map.

  The display control unit 50 causes the display unit 70 to display data representing the depth states in the plurality of object regions in the depth map, which are respectively processed by the depth map processing unit 20. At that time, the data is displayed in such a manner that the positional relationship between the objects in the depth map can be intuitively recognized.

  Specifically, the display control unit 50 displays a list of the average value, the minimum value, and the maximum value of the depth values in each object area in the depth map in a graph for each layer. The display control unit 50 may display the depth value in each object area as a histogram.

  FIG. 18 is a diagram illustrating an example of a graph displaying the depth state for each layer. The vertical axis of the graph in FIG. 18 represents the layer number, and the horizontal axis represents the depth value scale. In this example, the number of layers in the 2D input image is 16 (layer 0 to layer 15). In FIG. 18, the layer 0 object appears most retracted when viewed from the observer, and the layer 12 object appears most popped out.

  In each layer, both ends of arrows (see symbol r) indicate the minimum value (see symbol p) and the maximum value (see symbol q) of the depth value in the object area of each layer. That is, the arrows (see symbol r) indicate the amplitude and dynamic range of the depth value in the object area of each layer. The line connecting the arrows is a line connecting the average values of the depth values (see symbol o) in the object areas of the layers.

  The user observes a graph indicating the depth states of all layers displayed on the display unit 70, and edits or processes the depth map individually for each layer. Note that the user may input numerical values such as a gain value and an offset value from the keyboard, or may directly operate the displayed graph. When the state of the depth map processed by the depth map processing unit 20 changes, the display control unit 50 reflects the change on the display unit 70.

  The display control unit 50 may display the thumbnail image of the object of the layer in the vicinity of the identification symbol (number in FIG. 18) of each layer. As a result, the data indicating the depth state of each layer can be intuitively associated with the object image. The display control unit 50 may cause the display unit 70 to display four types of images: a graph indicating the depth state of all layers, a combined depth map, a 2D input image, and a 3D input image.

  FIG. 19 is a diagram illustrating an example of displaying the depth state of a layer as a histogram. This histogram is obtained by replacing one arrow in FIG. In the arrow display of FIG. 18, the user cannot recognize the distribution of the depth value in the object area of each layer, but can also recognize the distribution in the histogram display. Similarly to the arrow display of FIG. 18, the minimum value (see symbol t), maximum value (see symbol u), and average value (see symbol s) of the depth value in the object area of the layer can be displayed in the histogram display. Further, a histogram as shown in FIG. 19 may be displayed side by side for each layer.

  Hereinafter, a method of calculating the average value, the maximum value, and the minimum value of the depth values in the object area of each layer will be described. First, a method for calculating the average value of the depth values in the object area will be described. First, the depth map processing unit 20 counts the total number of pixels in the object area. Next, the sum of depth values for all pixels in the object area is calculated. The sum calculated at the end is divided by the total number of pixels counted.

  Next, a method for calculating the minimum value of the depth value in the object area will be described. First, the depth map processing unit 20 writes the maximum value that can be stored in the work area as an initial value in the work area for storing the minimum value of the depth value. Next, the depth values for all the pixels in the object area are referred to in order. The depth value of the referenced pixel is compared with the value in the work area at that time. If the former value is smaller than the latter value, the former value is overwritten in the work area. If the former value is greater than or equal to the latter value, this overwriting is skipped. This process is repeated for all pixels in the object area. The value stored in the work area at the time when the processing is completed for all pixels is determined as the minimum value of the depth value in the object area.

  Finally, a method for calculating the maximum depth value in the object area will be described. First, the depth map processing unit 20 writes the minimum value that can be stored in the work area as an initial value in the work area for storing the maximum value of the depth value. Next, the depth values for all the pixels in the object area are referred to in order. The depth value of the referenced pixel is compared with the value in the work area at that time. If the former value is larger than the latter value, the former value is overwritten in the work area. If the former value is less than or equal to the latter value, this overwriting is skipped. This process is repeated for all pixels in the object area. The value stored in the work area at the time when the processing is completed for all pixels is determined as the maximum depth value in the object area.

  As described above, according to the present embodiment, when generating a 3D image from a 2D image, a high-quality 3D image can be generated while reducing the user's trouble. That is, a depth map reflecting the scene structure can be generated by automatic processing by estimating the scene structure by the depth map generation unit 10 and changing the composition ratio of a plurality of basic depth models. Further, the depth map processing unit 20 reflecting the user's instruction processes the depth map individually for each object in the depth map. As a result, a high-quality 3D image can be generated, and the amount of work can be greatly reduced as compared to a case where a user generates a 3D image of similar quality from the 2D image from scratch.

  The depth map used in the 2D3D conversion shown in FIG. 1 holds the depth value for the entire screen. By adjusting the gain or offset with respect to the depth map, the depth feeling of the entire screen can be freely adjusted. Further, by adding gradation to the depth map, the inclination of the depth degree of the entire screen can be freely adjusted.

  However, generally, there are a plurality of objects in an image. In the example of the present specification, there are a person, a tree, and a background. If the depth map is gain-adjusted to enhance the depth feeling of the person portion to adjust the depth feeling of the person portion, the depth feeling of trees and backgrounds other than the person will also change at the same time. Therefore, it is difficult to independently adjust only the sense of depth of the person portion. Also, if you adjust the depth map to emphasize the depth of the tree by adjusting the depth map to adjust the depth of the tree, the depth of the person other than the tree and the background will change at the same time. End up. Therefore, it is difficult to independently adjust only the sense of depth of the tree portion. Further, if gradation is added to the depth map in order to incline the depth of the background portion, the depth of the person other than the background and the portion of the tree is also inclined at the same time. Therefore, it is difficult to incline only the depth of the background portion.

  On the other hand, according to the image editing system 500 according to the embodiment, the unevenness degree and the inclination of each object in the image can be freely changed independently. Therefore, the user can freely and independently control the perspective of the 3D image generated based on the depth map used in the 2D3D conversion according to the embodiment for each individual object in the image. Therefore, a high-quality 3D image can be generated.

  In addition, by displaying the data representing the depth state in multiple object areas in the depth map in a graph, the depth relationship between multiple layers can be easily grasped, greatly improving the efficiency of depth layer processing. it can. In other words, since it becomes easy to grasp the reference for processing the depth map that holds the disparity information that is the reference for generating the disparity, it is possible to give an appropriate disparity to the generated image, and to determine the object context of the generated 3D image. Can be adjusted appropriately. Further, it is possible to grasp the depth value for each layer without grasping the object area of the depth map by the simultaneous observation of the depth map and the layer, which is inherently difficult. Further, the depth value for each layer can be clearly identified without being confused by local shading due to embossing or gradation.

  The present invention has been described based on the embodiments. The embodiments are exemplifications, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are within the scope of the present invention. .

  500 image editing system, 100 image processing apparatus, 10 depth map generation unit, 11 upper screen high frequency component evaluation unit, 12 lower screen high frequency component evaluation unit, 13 synthesis ratio determination unit, 14 frame memory for first basic depth model, 15 Frame memory for 2nd basic depth model, 16 Frame memory for 3rd basic depth model, 17 Composition part, 18 Adder part, 20 Depth map processing part, 30 3D image generation part, 40 Operation reception part, 50 Display control part, 200 console terminal device, 60 operation unit, 70 display unit.

Claims (6)

A depth map generating unit that generates a depth map of the input image based on the input image and the depth model;
A depth map processing unit that processes the depth map generated by the depth map generation unit for each of a plurality of regions in a screen specified by a plurality of mask patterns set externally;
Based on the input image and the depth map processed by the depth map processing unit, an image generation unit that generates an image of another viewpoint;
A display control unit that displays the data representing the state of depth in a plurality of designated areas in the depth map, each processed by the depth map processing unit, in a graph on a display;
An image processing apparatus comprising:   The image processing apparatus according to claim 1, wherein the display control unit displays an average value, a minimum value, and a maximum value of the depth values in each designated area in the depth map in a graph.   The image processing apparatus according to claim 1, wherein the display control unit displays a depth value in each designated region in the depth map as a histogram. The depth map generation unit generates a depth model to be used for depth map generation by combining a plurality of basic depth models,
The image processing apparatus according to claim 1, wherein the depth map generation unit changes a synthesis ratio of a plurality of basic depth models according to a scene structure of an input image. Generating a depth map of the input image based on the input image and the depth model;
Processing the generated depth map for each of a plurality of areas in a screen specified by a plurality of externally set mask patterns;
A step of displaying data representing depth states in a plurality of designated areas in the depth map on the display in a graph, each processed;
Generating a different viewpoint image based on the input image and the processed depth map;
An image processing method comprising: A process of generating a depth map of the input image based on the input image and the depth model;
Processing the generated depth map for each of a plurality of areas in the screen specified by a plurality of mask patterns set externally;
Processing each of the data that represents the state of the depth in a plurality of designated areas in the depth map processed in a graph on the display,
Based on the input image and the processed depth map, processing for generating an image of another viewpoint;
An image processing program for causing a computer to execute.