JP5516199B2 - Image processing apparatus, image processing method, projection apparatus, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, a projection apparatus, and a program for creating a three-dimensional image from a two-dimensional image.

  In order to create an uncomfortable 3D image from the 2D image, for example, a horizontal contour component of the 2D image is detected, and a left eye image and a right eye image are detected according to the horizontal maximum point of the detected contour component. The technology that has been made to create is considered. (For example, Patent Document 1)

JP 2004-320189 A

  In the technique described in the above patent document, a virtual depth position is set according to a position in a rectangular image, or a distance to a subject in the image is set by detecting a horizontal contour component of the image. A left-eye image and a right-eye image are created.

  However, in many images that are actually photographed, it is difficult to obtain a contour component in the horizontal direction that is photographed with a composition other than a pre-patterned composition or that is in focus but has a clear sense of distance. There seems to be many things. Accordingly, the depth position set by the position in the image or the content of the distance of the contour portion is not always correct, and on the contrary, there is a possibility that a 3D image with a sense of incongruity may be obtained.

  The present invention has been made in view of the above circumstances, and the object of the present invention is to create a three-dimensional image that has no sense of incongruity and is a more natural three-dimensional description from an existing two-dimensional image. An image processing apparatus, an image processing method, a projection apparatus, and a program are provided.

The invention according to claim 1 is an image processing apparatus, wherein an image input unit that inputs a two-dimensional image in which a plurality of subjects each including a shadow are reflected, and a two-dimensional image input by the image input unit The subject part and its shadow part are extracted, and the distance to the subject part is estimated according to the distance from the closest subject part to the lower side of the image from the corresponding subject part of the plurality of shadow parts extracted by the extraction part A distance estimation unit, and an image creation unit that creates an image for both left and right eyes in which the subject part and the shadow part are shifted based on the distance obtained by the distance estimation unit are provided.

  According to a second aspect of the present invention, in the first aspect of the invention, the two-dimensional image input by the image input unit includes shooting condition information including a shooting angle of view, and the distance estimation unit includes the plurality of shadows. In addition to the positional relationship of the portions, the distance to the corresponding subject portion is estimated by referring to the information of the shooting angle of view included in the shooting condition information.

  According to a third aspect of the present invention, in the first aspect of the invention, the distance estimation unit estimates a distance to a corresponding subject portion based on a relationship between positions closest to the subject in each of the plurality of shadow portions. It is characterized by doing.

The invention according to claim 4 is an image processing method, wherein an image input step of inputting a two-dimensional image in which a plurality of subjects each including a shadow is reflected, and the two-dimensional image input in the image input step An extraction step of extracting the subject portion and its shadow portion, and estimating the distance to the subject portion according to the distance from the closest position to the lower side of the image from the corresponding subject portion of the plurality of shadow portions extracted in the extraction step The method includes a distance estimation step and an image creation step of creating an image for both left and right eyes in which the subject portion and the shadow portion are shifted based on the distance obtained in the distance estimation step.

The invention according to claim 5 is a projection apparatus, wherein an image input unit that inputs a two-dimensional image in which a plurality of subjects each including a shadow are reflected, and a two-dimensional image input by the image input unit An extraction unit that extracts a subject part and its shadow part, and a distance that estimates the distance to the subject part according to the distance from the closest position to the lower side of the image from the corresponding subject part of the plurality of shadow parts extracted by the extraction unit An estimation unit, an image creation unit that creates images for left and right eyes that are shifted from the subject part and the shadow part based on the distance obtained by the distance estimation unit, and for both left and right eyes created by the image creation unit And a projection unit for projecting the image.

  The invention described in claim 6 is the invention described in claim 5, further comprising a distance acquisition unit that acquires a distance to a projection target on which the projection unit projects an image, and the image creation unit includes the distance It is characterized in that an image for left and right eyes is generated by shifting the corresponding part based on the distance obtained by the estimation unit and the distance obtained by the distance obtaining unit.

The invention according to claim 7 is a program executed by a computer built in the image processing apparatus, and an image input unit that inputs a two-dimensional image in which a plurality of subjects each including a shadow are reflected. An extraction unit for extracting the subject part and its shadow part in the two-dimensional image input by the image input unit, and a distance from a position closest to the corresponding subject part of the plurality of shadow parts extracted by the extraction unit to the lower side of the image And a distance estimation unit that estimates the distance to the subject part, and an image creation unit that creates images for the left and right eyes with the subject part and the shadow part shifted based on the distance obtained by the distance estimation unit. Let

  According to the present invention, when a plurality of subjects in an existing two-dimensional image are reflected together with shadows, the positional relationship of each subject is reproduced, and there is no sense of incongruity, resulting in a more natural three-dimensional depiction. A three-dimensional image can be created.

The block diagram which shows the structure of the functional circuit of the data projector apparatus which concerns on one Embodiment of this invention. 3 is a flowchart showing processing details of stereoscopic image projection using two-dimensional still image data according to the embodiment. The figure which shows an example of the still image which concerns on the same embodiment. The figure which shows an example of the still image which concerns on the same embodiment. The figure which shows the relationship between the imaging distance p of the to-be-photographed object which concerns on the embodiment, the coefficient k, the imaging | photography angle of view, and the distance D from the image lower side. The figure explaining the concept of the shift amount based on parallax for providing the perspective which concerns on the embodiment. 6 is a timing chart showing projection timings in the liquid crystal shutter glasses method according to the embodiment.

  An embodiment in which the present invention is applied to a data projector apparatus of DLP (Digital Light Processing) (registered trademark) system that projects a stereoscopic image will be described below with reference to the drawings.

  As for the projection of the stereoscopic image, for example, an image of a liquid crystal shutter glasses type is created. This liquid crystal shutter glasses method is also called a field sequential method or an active stereo method and projects an image for one frame divided into two fields. The image for the left eye in the first field and the image for the right eye in the second field. Project an image.

  For example, infrared light outside the visible light range is projected on the image only in the first field. The user wears liquid crystal shutter glasses for this projection. On the liquid crystal shutter glasses side, the infrared light is detected by a sensor, and the right eye side is shuttered in the first field, and the left eye side is shuttered in the second field. Therefore, the user can see only the image for the left eye with the left eye and the image for the right eye separately with the right eye without being aware of it.

  It should be noted that liquid crystal shutter glasses used in the liquid crystal shutter glasses themselves can be used as they are for commercially available 3D television receivers, and in this embodiment, description of the configuration, operation, etc. is omitted.

FIG. 1 is a block diagram showing a schematic configuration of a functional circuit of a data projector device 10 according to the present embodiment.
Reference numeral 11 denotes an input unit. The input unit 11 includes, for example, a pin jack (RCA) type video input terminal, a D-sub 15 type RGB input terminal, an HDMI (High-Definition Multimedia Interface) standard image / audio input terminal, and a USB (Universal Serial Bus). An analog image signal and audio signal from an external device connected by wire are input via a connector, and digitized by performing A / D conversion and other predetermined processing as front-end processing.

  The image signals of various standards input and digitized by the input unit 11 are sent to the projection image processing unit 12 via the system bus SB.

  The projection image processing unit 12 unifies the input image signal into an image signal of a predetermined format suitable for projection, writes the image signal to the display video memory 13 as appropriate, and then writes the written image signal from the video memory 13. Read out and send to the projected image drive unit 14.

  At this time, symbol data indicating various operation states for OSD (On Screen Display) and character data such as a guide message are also superimposed on the image signal in the video memory 13 by the projection image processing unit 12 as necessary. The image signal is read and sent to the projection image drive unit 14.

  The projection image drive unit 14 multiplies a frame rate according to a predetermined format, for example, 120 [frames / second], the number of color component divisions, and the number of display gradations, in accordance with the transmitted image signal. The micromirror element 15 that is a spatial light modulation element (SLM) is driven to display by high-speed time-division driving.

  The micromirror element 15 performs display operation by individually turning on / off each tilt angle of a plurality of micromirrors arranged in an array, for example, XGA (horizontal 1024 pixels × vertical 768 pixels) at high speed. Then, an optical image is formed by the reflected light.

  On the other hand, R, G, B primary color lights are emitted cyclically from the light source unit 16 in a time-sharing manner. The primary color light from the light source unit 16 is totally reflected by the mirror 17 and applied to the micromirror element 15.

  Then, an optical image is formed by the reflected light from the micromirror element 15, and the formed optical image is projected and displayed on a screen (not shown) to be projected via the projection lens unit 18.

  The projection lens unit 18 has a zoom lens and a focus lens inside, and can change the projection angle of view and the focus position by moving each lens.

  The light source unit 16 includes a light emitting diode (hereinafter referred to as “Ir-LED”) 19 that emits infrared light (Ir), a light emitting diode (hereinafter referred to as “R-LED”) 20 that emits red (R) light, and green ( G) a light emitting diode (hereinafter referred to as “G-LED”) 21 that emits light, and a light emitting diode (hereinafter referred to as “B-LED”) 22 that emits blue (B) light.

  Unlike the other R-LED 20, G-LED 21, and B-LED 22, the Ir-LED 19 is driven to emit light by switching on / off in units of fields when projecting a stereoscopic image. Infrared light emitted from the Ir-LED 19 is transmitted through the dichroic mirror 23, converted into a luminous flux having a substantially uniform luminance distribution by the integrator 24, and then sent to the mirror 17.

  The red light emitted from the R-LED 20 is reflected by the dichroic mirror 25 and then by the dichroic mirror 23, and is sent to the mirror 17 after the luminance distribution is made substantially uniform by the integrator 24.

  The green light emitted from the G-LED 21 is reflected by the dichroic mirror 26, then passes through the dichroic mirror 25, is then reflected by the dichroic mirror 23, and is sent to the mirror 17 through the integrator 24.

The blue light emitted from the B-LED 22 is reflected by the mirror 27, passes through the dichroic mirrors 26 and 25, is then reflected by the dichroic mirror 23, and is sent to the mirror 17 through the integrator 24.
The dichroic mirror 23 transmits infrared light while reflecting red light, green light, and blue light. The dichroic mirror 25 reflects red light and transmits green light and blue light. The dichroic mirror 26 reflects green light while transmitting blue light.

  The projection light drive unit 28 controls the light emission timing of each of the LEDs 19 to 22 of the light source unit 16 and the waveform of the drive signal. The projection light drive unit 28 controls the light emission operation of the LEDs 19 to 22 according to the timing of the image data given from the projection image drive unit 14 and the control of the CPU 29 described later.

  The CPU 29 controls all the operations of the above circuits. The CPU 29 is directly connected to the main memory 30 and the program memory 31. The main memory 30 is composed of a DRAM and functions as a work memory for the CPU 29. The program memory 31 is composed of an electrically rewritable nonvolatile memory, and stores an operation program executed by the CPU 29, various fixed data, and the like.

  The CPU 29 controls the data projector device 10 by reading out the operation program and various fixed data stored in the program memory 31, expanding and storing them in the main memory 30, and executing the program. Control.

The CPU 29 executes various projection operations in accordance with key operation signals from the operation unit 32.
The operation unit 32 includes a key operation unit provided in the main body of the data projector device 10 and a laser light receiving unit that receives infrared light from a remote controller (not shown) dedicated to the data projector device 10. The operation unit 32 directly outputs to the CPU 29 a key operation signal based on a key operated by a user with a key operation unit of the main body or a remote controller.

  Specifically, the power key, input switch key, focus up / down key, zoom up / down key, menu key, cursor ("↑" "↓") “←” “→”) key, set key, cancel key, keystone correction key, and the like.

The CPU 29 is further connected to the image memory 33, the image processing unit 34, and the sound processing unit 35 via the system bus SB.
The image memory 33 has a storage capacity capable of storing a plurality of various image data sent via the input unit 11. Under the control of the CPU 29, the image processing unit 34 performs image processing including image data generation for stereoscopic image projection on the image data stored in the image memory 33.

  The sound processing unit 35 includes a sound source circuit such as a PCM sound source, converts the sound data given during the projection operation into an analog signal, drives the speaker unit 36 to emit a loud sound, or generates a beep sound or the like as necessary.

Next, the operation of the above embodiment will be described.
As described above, the projection image processing unit 12 creates an image to be displayed on the micromirror element 15 using the video memory 13, and the projection image driving unit 14 displays the created image on the micromirror element 15. The projection light drive unit 28 drives the LEDs 19 to 22 to emit light in accordance with the display on the element 15.

The projection image processing unit 12, the video memory 13, the projection image driving unit 14, and the projection light driving unit 28 all operate under the control of the CPU 29. The CPU 29 reads out an operation program, fixed data, and the like stored in the program memory 31 including the processing described below, and executes the control processing after developing it in the main memory 30.
Various processes for converting the image data stored in the image memory 33 into stereoscopic image data are executed by the image processing unit 34 under the control of the CPU 29.

  FIG. 2 is a flowchart showing a series of processing contents when a stereoscopic image is projected using a plurality of two-dimensional still image data and moving image data input from the input unit 11 and stored in the image memory 33. is there.

  At the beginning of the processing, it is accepted that a user selects still image data as a background from among a plurality of still image data stored in the image memory 33 (step S101).

  In this case, for example, thumbnail images of all the still image data stored in the image memory 33 are created, and after creating a list image thereof, a guide message for selecting one of the thumbnail images is projected.

When the user operates the cursor key on the operation unit 32 with respect to the projection content, the position of the selected image in the list image is changed according to the operation content, and the image selected when the set key is operated is selected. Determine the background image.
Next, it is accepted that the part image of the subject to be projected on the background image is selected by the user from the plurality of moving image data stored in the image memory 33 (step S102).

  In this case, thumbnail images are created using still images located at the heads of all the moving image data stored in the image memory 33, a list of those images is created, and one of them is selected. Project with a guide message.

  When the user operates the cursor key on the operation unit 32 with respect to the projection content, the position of the selected image in the list image is changed according to the operation content, and the image selected when the set key is operated is selected. Decide on a part image.

  When the selection of the background still image and the moving image to be combined with the background still has been completed, the subject and its shadow portion in the image are extracted from the still image data selected as the background (step S103). .

In extracting the subject portion, for example, processing such as image contour extraction is performed, and the shape of the region surrounded by the contour is recognized from the surrounding image and then cut out.
Further, in extracting a shadow portion for a subject, after extracting the color component images of R, G, and B, after extracting the shadow area (area where the light from the light source is not directly hit) and the boundary (shadow shadow) of the non-shadow area (Boundary) extraction processing is performed, and a shadow portion is recognized from each extraction result. The shadow portion recognition process is a known technique as described in, for example, Japanese Patent Application Laid-Open No. 2007-272292, and therefore, detailed description thereof is omitted.

  Next, attention is paid only to the shadow portion extracted in the immediately preceding step S103 (step S104), and the shadow portion that is reflected together with the subject portion depending on whether there are a plurality of shadow portions that can be extracted corresponding to the subject portion. It is determined whether or not there is a plurality (step S105).

  Here, when it is determined that there are a plurality of shadow portions that can be extracted corresponding to the subject portion, among the plurality of optical lenses constituting the projection lens unit 18, from the focus lens driving position at that time, The distance to the eyelid screen at the in-focus position is detected by a lens drive motor (not shown) and a rotary encoder for detecting the rotation position (step S106).

Then, the positional relationship between the front and rear and the depth of the plurality of shadow portions extracted in step S104 is determined (step S107).
FIG. 3 shows a first example of a still image in which a plurality of persons are captured as subjects. In the figure, there are some people as subjects with shadows. When attention is paid to, for example, the persons S11 and S12, the closest points P11 and P12 with the subject are obtained for the extracted shadow portions H11 and H12, respectively. Then, distances D11 and D12 from the closest points P11 and P12 to the lower side of the image are obtained.

  Thus, it is estimated that the larger value of the distance D from the lower side of the image is located farther in relation to the depth than the shorter one. In this case, the distances D11 and D12 may be represented by a ratio of the number of pixels with respect to the distance to the vertical image size (number of pixels) of the still image data.

  FIG. 4 shows a second example of a still image in which mainly two persons are captured as subjects. Focusing on two persons S21 and S22 with shadows in the figure, shadow portions H21 and H22 can be extracted. In this case, the person S22 is in a jumped state and is away from the shadow portion H22.

  For each of the shadow portions H21 and H22, the closest points P21 and P22 with the corresponding subject are obtained. Then, distances D21 and D22 from the closest points P21 and P22 to the lower side of the image are obtained.

  Next, it is determined whether or not shooting condition information is added to the still image data (step S108).

  Normally, if the still image data conforms to the DCF (Design rule for Camera File system) standard, each information of the focal length at the time of shooting and the size of the image sensor is taken together with the shooting date, shutter speed, aperture value, etc. It is attached as condition information, and the shooting angle of view can be calculated from this information.

  Therefore, if it is determined in step S108 that the shooting condition information is added to the still image data, the three-dimensional image of the subject linked to the shadow has a front-rear relationship corresponding to the depth position of each subject in the image. The shift amount is determined in consideration of the shooting angle of view (step S109).

  FIG. 5 is a diagram showing the relationship between the shooting distance p of each subject, the coefficient k, the shooting angle of view, and the distance D from the lower side of the image. After estimating the distance p to the subject based on the relationship shown in this figure, a shift amount corresponding to the distance p is determined.

  When the size of the subject S does not change, the distance to the subject and the apparent size are inversely proportional. Therefore, when the distance to the subject is s and the magnification is r, the distance p to the subject is calculated as s / r, and the shift amounts of the left and right eyes are calculated from the calculated distance p.

FIG. 6 illustrates an expression corresponding to the position of the subject S when the position of the background image is the position of the screen SC to be projected.
FIG. 6A shows a representation of the subject S that is farther from the screen SC. FIG. 6B shows a representation of the subject S located closer to the screen SC.

If the right-eye image is shifted to the right, the left-eye image shifted to the left is a positive value, the right-eye image is shifted to the left, and the left-eye image is shifted to the right, the shift amount is expressed by the following equation. Is done. That is,
w = d · (1− (s / p)) (1)
(W: Shift amount,
d: the distance between the center positions of the left and right eyes,
s: distance to the screen,
p: Distance to the subject. )
In the above formula (1), the distance s to the screen SC has already been acquired in step S106.

  If it is determined in step S108 that the shooting condition information is not added to the still image data, the three-dimensional image of the subject linked to the shadow has a front-to-back relationship corresponding to the depth position of each subject in the image. The shift amount is determined without considering the shooting angle of view (step S110).

  Based on the shift amounts calculated in this way, the moving image data for the left and right eyes is created by further combining the moving image data as the part image selected in step S102 (step S111).

  The moving image data composed of the left-eye image and the right-eye image created in this way is converted into image data by the image processing unit 34 in accordance with the format defined by the liquid crystal shutter glasses method described above, and is stored in the image memory 33 (step) S112).

  Next, the projection of a stereoscopic image by the liquid crystal shutter glasses method is executed based on the moving image data of the left-eye image and the right-eye image stored in the image memory 33 (step S113). A series of processes up to the projection of the stereoscopic image based on the image data is completed.

  FIG. 7 shows each projection timing in the liquid crystal shutter glasses method. As described above, in the liquid crystal shutter glasses method, an image for one frame is divided into two fields, and a left-eye image is projected in the first field and a right-eye image is projected in the second field.

  FIG. 7A shows a light emission drive waveform of the Ir-LED 19 that is lit only in the first field in one field and two fields. The infrared light emitted from the Ir-LED 19 is superimposed on the image light and emitted from the projection lens unit 18, so that an infrared sensor provided on the liquid crystal shutter glasses (not shown) used by the user emits the infrared light. Then, switching between the left-eye image and the right-eye image is performed.

  The R-LED 20, the G-LED 21, and the B-LED 22 emit light sequentially in a time division manner in each field as shown in FIGS. 7B to 7D.

  In synchronization with the light emission of these primary color LEDs 20 to 22, the micromirror element 15 causes the red (R) image, green (G) image, and blue (B) image for the left eye in the first field as shown in FIG. In the second field, a red (R) image, a green (G) image, and a blue (B) image for the right eye are cyclically displayed, so that a light image is formed by each reflected light thereof, and the projection lens Projected onto the screen SC by the unit 18.

  Further, when it is determined in step S105 that there are not a plurality of shadow portions reflected together with the subject portion, the three-dimensional projection is not performed on each subject in the still image as the background, and the process directly proceeds to step S111. In step S102, a moving image is created by simply synthesizing the moving image data as the part image selected in step S102.

  The created moving image data is converted into image data by the image processing unit 34 in accordance with the reproduction format and stored in the image memory 33 (step S112).

  Next, the projection is executed based on the moving image data stored in the image memory 33 (step S113), and the series of processes from the selection of the image data to the projection of the synthesized moving image based on the selected image data is completed. .

  As described above in detail, according to the present embodiment, when a plurality of subjects in an existing two-dimensional image are reflected together with shadows, the positional relationship of each subject is reproduced, and there is no sense of incongruity and is more natural. It is possible to create a three-dimensional image that represents a three-dimensional effect.

  In addition, in the above embodiment, when the still image data includes shooting condition information including the shooting angle of view, the distance to the subject is estimated in consideration of the shooting angle of view. Thus, for example, the distance to the subject is estimated so that the position of the subject in the still image photographed using the telephoto lens is assumed to be farther than the display position in the still image photographed using the wide-angle lens. Since the estimation is made more accurately according to the angle of view, it is possible to project a stereoscopic image with a more accurate description and a more natural description.

  Furthermore, in the above embodiment, since the distance is estimated based on the position of the most corresponding subject in the shadow portion, it is always accurate without being influenced by the direction of the light source and the subject such as follow light and backlight. In addition, the positional relationship of the subject and the shooting distance can be accurately estimated.

  In the above embodiment, the case where the present invention is applied to a projection apparatus will be described. The distance to the projection target screen is acquired, and the stereoscopic effect to be added to the image is adjusted based on the acquired distance.

  As a result, the stereoscopic effect of the image is very natural for the user who appreciates the projection content, and the image can be rendered without any sense of incongruity.

  In the above embodiment, the case where the liquid crystal shutter glasses method is employed as the stereoscopic image projection method has been described, but the present invention is not limited to this, and the dual stream format, the side-by-side format, the top and bottom format, the field sequential format, The same can be applied to the case of projecting a stereoscopic image such as an interlace format.

  Although the above embodiment has been described with respect to a case where the present invention is applied to a DLP (registered trademark) projector device, the present invention is not limited to the projection method, and a transmissive color liquid crystal panel is used. Various applications such as a monochrome liquid crystal panel in which light sources of a plurality of colors are driven in a time-sharing manner are applicable.

  Furthermore, the present invention is not limited to a projection apparatus, and an electronic apparatus having various image display functions for generating and displaying a stereoscopic image, such as a personal computer, a television receiver, a digital camera, a DVD (Digital Versatile Disc) player, and an electronic book. It can also be applied to mobile phone terminals, media players, and the like.

  In addition, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention in the implementation stage. Further, the functions executed in the above-described embodiments may be combined as appropriate as possible. The above-described embodiment includes various stages, and various inventions can be extracted by an appropriate combination of a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, if the effect is obtained, a configuration from which the constituent requirements are deleted can be extracted as an invention.

  DESCRIPTION OF SYMBOLS 10 ... Data projector apparatus, 11 ... Input part, 12 ... Projection image process part, 13 ... Video memory, 14 ... Projection image drive part, 15 ... Micromirror element, 16 ... Light source part, 17 ... Mirror, 18 ... Projection lens unit 19 ... Ir-LED, 20 ... R-LED, 21 ... G-LED, 22 ... B-LED, 23 ... Dichroic mirror, 24 ... Integrator, 25,26 ... Dichroic mirror, 27 ... Mirror, 28 ... Projection light drive , 29 ... CPU, 30 ... main memory, 31 ... program memory, 32 ... operation unit, 33 ... image memory, 34 ... image processing unit, 35 ... audio processing unit, 36 ... speaker unit, SB ... system bus.

Claims (7)

An image input unit for inputting a two-dimensional image in which a plurality of subjects each including a shadow are reflected;
An extraction unit for extracting a subject part and its shadow part in the two-dimensional image input by the image input unit;
A distance estimation unit that estimates the distance to the subject portion according to the distance from the closest position to the lower side of the image from the corresponding subject portion of the plurality of shadow portions extracted by the extraction unit;
An image processing apparatus, comprising: an image creating unit that creates an image for both left and right eyes in which the subject part and the shadow part are shifted based on the distance obtained by the distance estimating unit. The two-dimensional image input by the image input unit includes shooting condition information including a shooting angle of view,
The distance estimation unit estimates a distance to a corresponding subject portion with reference to information on a shooting angle of view included in the shooting condition information in addition to the positional relationship of the plurality of shadow portions. The image processing apparatus according to 1.   The image processing apparatus according to claim 1, wherein the distance estimation unit estimates a distance to a corresponding subject portion based on a relationship between positions closest to the subject in each of the plurality of shadow portions. An image input step of inputting a two-dimensional image in which a plurality of subjects each including a shadow are reflected;
An extraction step of extracting the subject portion and its shadow portion in the two-dimensional image input in the image input step;
A distance estimation step of estimating the distance to the subject portion according to the distance from the closest position to the lower side of the image from the corresponding subject portion of the plurality of shadow portions extracted in the extraction step;
An image processing method comprising: an image creation step of creating an image for left and right eyes, in which the subject portion and the shadow portion are shifted based on the distance obtained in the distance estimation step. An image input unit for inputting a two-dimensional image in which a plurality of subjects each including a shadow are reflected;
An extraction unit for extracting a subject part and its shadow part in the two-dimensional image input by the image input unit;
A distance estimation unit that estimates the distance to the subject portion according to the distance from the closest position to the lower side of the image from the corresponding subject portion of the plurality of shadow portions extracted by the extraction unit;
An image creating unit that creates an image for both left and right eyes in which the subject part and the shadow part are shifted based on the distance obtained by the distance estimating unit;
A projection apparatus comprising: a projection unit configured to project an image for both left and right eyes created by the image creation unit. The projection unit further includes a distance acquisition unit that acquires a distance to a projection target on which an image is projected,
The image creating unit according to claim 5, characterized in that to create the image for the right and left eyes shifted that part based on the distance obtained by the distance and the distance obtaining unit obtained by the distance estimating unit Projection device. A program executed by a computer built in the image processing apparatus,
The program
An image input unit for inputting a two-dimensional image in which a plurality of subjects each including a shadow are reflected;
An extraction unit for extracting a subject part and its shadow part in the two-dimensional image input by the image input unit;
The distance estimation unit for estimating the distance to the subject part according to the distance from the closest position to the lower side of the image from the corresponding subject part of the plurality of shadow parts extracted by the extraction unit, and the distance obtained by the distance estimation unit A program for functioning as an image creation unit for creating an image for both left and right eyes, in which the subject part and the shadow part are shifted based on each other.

Images