WO2009151292A2 - 영상 변환 방법 및 장치 - Google Patents
영상 변환 방법 및 장치 Download PDFInfo
- Publication number
- WO2009151292A2 WO2009151292A2 PCT/KR2009/003151 KR2009003151W WO2009151292A2 WO 2009151292 A2 WO2009151292 A2 WO 2009151292A2 KR 2009003151 W KR2009003151 W KR 2009003151W WO 2009151292 A2 WO2009151292 A2 WO 2009151292A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- information
- depth information
- original image
- stereoscopic
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/261—Image signal generators with monoscopic-to-stereoscopic image conversion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/275—Image signal generators from 3D object models, e.g. computer-generated stereoscopic image signals
Definitions
- the present invention relates to image conversion, and more particularly, to a method and apparatus for converting a two-dimensional image into a three-dimensional image (ie, a stereoscopic image).
- the parallax is generated on the retina. Due to this parallax, images of the two eyes are synthesized into one from the cerebrum to recognize the stereoscopic object.
- images shown in photographs, videos, and TV are flat images without a three-dimensional effect because they are artificially created using video cameras or still cameras.
- a method of generating a stereoscopic image by acquiring a plurality of images of a target object using a plurality of cameras and then combining them is present.
- this is a method of generating a stereoscopic image according to special conditions at the time of capturing, and there is a problem that it is virtually impossible to convert the stereoscopic image to a stereoscopic image without recapturing a previously photographed movie.
- the present invention relates to an image conversion method and apparatus capable of converting a two-dimensional image to a three-dimensional image.
- the present invention relates to an image conversion method and apparatus capable of expressing both intaglio and emboss in a three-dimensional image using a zero point.
- a method for converting a two-dimensional image into a stereoscopic image by an image conversion apparatus, and a recording medium on which a program for performing the method is recorded is recorded.
- a method for converting a two-dimensional image into a stereoscopic image by the image conversion apparatus comprising: receiving and setting depth information about the original image; Dividing the original image into partial objects, and setting stereoscopic information for each partial object; Generating a first image by moving the original image using the stereoscopic information; Receiving and setting a zero point for the original image; Generating a second image by moving the original image using the zero point; And generating a stereoscopic image by synthesizing the first image and the second image.
- the stereoscopic information includes depth information and contour information on each partial object, and the contour information may be area information of partial objects having the same depth information.
- the zero point is depth information corresponding to the display plane.
- Setting stereoscopic information for each partial object comprises: setting a center object in the original image; Dividing a plurality of partial objects in order from the center object to a peripheral object, and setting depth information on the divided partial objects; And setting the contour information for the partial objects having the same depth information.
- the setting of the stereoscopic information may include: separating partial objects at an nth time point, and setting depth information on the divided partial objects; Generating an nth layer to include partial objects corresponding to the depth information set at the nth time point, wherein the nth layer includes partial objects having the same depth information, and wherein the nth layer is n- Stereoscopic information about partial entities corresponding to one layer is included, and n is a natural number.
- the first image may be generated by moving the original image in a horizontal direction determined according to stereoscopic information.
- the generating of the first image may include moving the partial objects included in the n-th layer in a horizontal direction corresponding to the set depth information, wherein the depth information is different from the depth information corresponding to the n-th layer. Value.
- the generating of the second image may include calculating movement information using the zero point and the depth information; And generating the second image by moving the original image in a horizontal direction determined according to the movement information.
- the moving direction of the original image for generating the second image is different from the moving direction of the original image for generating the first image.
- the method may further include dividing the video in each frame unit and sequentially storing each divided frame as a still image. .
- n-th frame Loading stereo information corresponding to an n-th frame when the original image is not the first frame; Comparing the n-th frame with the n-th frame to extract a changed region; And modifying stereoscopic information corresponding to the n-th frame corresponding to the changed area, wherein n is a natural number, and the n-th frame is a frame corresponding to a target still image to be converted at the present time.
- the n-th frame is a frame corresponding to the still image converted at a previous time point adjacent to the n-th frame.
- a program of instructions that can be executed by a digital processing apparatus is tangibly implemented to perform a method for converting a two-dimensional image into a stereoscopic image, and can be read by the digital processing apparatus.
- a recording medium recording a program, comprising: receiving and setting depth information of an original image; Dividing the original image into partial objects, and setting stereoscopic information for each partial object; Generating a first image by moving the original image using the stereoscopic information; Receiving and setting a zero point for the original image;
- a recording medium recording a program for synthesizing the first image and the second image to generate a stereoscopic image.
- an image conversion apparatus for converting a two-dimensional image to a stereoscopic image is provided.
- an image conversion apparatus for converting a two-dimensional image to a three-dimensional image, the apparatus comprising: a control unit for receiving and setting the depth information and the zero point for the original image; A setting unit for dividing the original image into partial entities and setting stereoscopic information for each partial entity; A generation unit for generating a first image by moving the original image using the stereoscopic information, and generating a second image by moving the original image using the zero point; And a synthesis unit configured to synthesize the first image and the second image to generate a stereoscopic image.
- the setting unit sets a central object in the original image, distinguishes a plurality of partial objects from the central object to a peripheral object, sets stereoscopic information to the divided partial objects, and partial objects having the same depth information. Contour information can be set for.
- the setting unit classifies the partial objects at an nth time point, sets depth information on the divided partial objects, and generates an nth layer including partial objects corresponding to the depth information set at the nth time point.
- the n-th layer arranges and outputs partial entities actually present at an n-th time point, and the n-th layer includes stereoscopic information about partial entities corresponding to the n-th layer, where n is a natural number.
- the generation unit may generate the first image by moving the original image in a horizontal direction determined according to the stereoscopic information.
- the generation unit moves the partial objects included in the n-th layer in a horizontal direction corresponding to the set depth information, wherein the depth information is a difference value from the depth information corresponding to the n-th layer.
- the generation unit may calculate movement information using the zero point and depth information of the original image, and generate the second image by moving the original image in a horizontal direction determined according to the movement information.
- the controller Before the controller receives the depth information and the zero point of the original image, the controller divides the video into frame units and converts each divided frame into a still image sequentially. Can be stored.
- the setting unit loads stereoscopic information corresponding to the n-th frame, extracts the changed region by comparing the n-th frame with the n-th frame, and changes the changed image.
- the stereoscopic information corresponding to the n-th frame is corrected according to an area, where n is a natural number, the n-th frame is a frame corresponding to a target still image to be converted at the present time, and the n-th frame is The frame corresponds to a still image converted at a previous time point adjacent to the n-th frame.
- the present invention has an effect that can be expressed both intaglio and embossed in the three-dimensional image by using a zero point.
- FIG. 1 is a view showing for explaining the principle of a three-dimensional image according to an embodiment of the present invention.
- FIG. 2 is a diagram schematically illustrating an image conversion according to an embodiment of the present invention.
- FIG. 3 is a block diagram illustrating an internal functional configuration of an image conversion apparatus according to an embodiment of the present invention.
- FIG. 4 is a flowchart illustrating a method of converting a 2D video into a 3D video according to an embodiment of the present invention.
- FIG. 5 is a flowchart illustrating a method of converting a two-dimensional image to a three-dimensional image according to an embodiment of the present invention.
- 6 to 9 illustrate images generated according to an embodiment of the present invention.
- FIG. 10 is a flowchart illustrating a three-dimensional image conversion method according to another embodiment of the present invention.
- FIG 11 and 12 illustrate screens provided by an image conversion apparatus according to an embodiment of the present invention.
- first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.
- FIG. 1 is a view illustrating a principle of a three-dimensional image according to an embodiment of the present invention.
- a two-dimensional image As a three-dimensional image (ie, a stereoscopic image), a plurality of parallax two-dimensional images are required.
- images having a parallax with respect to the original image 110 are generated using the original image 110.
- the first image having the parallax ie, the left eye image 120
- the second image that is, the right eye image 130
- a method of generating a first image and a second image having parallaxes from the original image 110 is described, but an implementation method is described.
- a three-dimensional image may be generated by synthesizing the original image 110 and the generated image by generating one image having a parallax with the original image 110.
- two-dimensional images are acquired through both eyes of a target object.
- the human brain may recognize a three-dimensional sense of a target object by synthesizing a two-dimensional image obtained through each eye.
- An image synthesized by generating a plurality of parallax images from an original image by using the same principle is called a stereoscopic image.
- a stereoscopic image is referred to as a stereoscopic image or a three-dimensional image.
- the 3D image may be generated by combining the left eye image 120 and the right eye image 130 of FIG. 1 according to a predetermined method.
- the 3D image is an image generated by synthesizing a plurality of still images having a parallax as shown in FIG. 1, and may be used to sense a three-dimensional effect using a specially manufactured equipment.
- FIG. 2 is a diagram schematically illustrating an image conversion according to an embodiment of the present invention.
- the image conversion apparatus 200 receives a two-dimensional image (ie, a still image) and a two-dimensional video.
- the image conversion apparatus 200 receives a two-dimensional image and a two-dimensional video, separates the plurality of partial objects, sets stereoscopic information according to each partial object, and converts the three-dimensional image and the three-dimensional video to output.
- the image conversion apparatus 200 divides the two-dimensional video into respective frame units, and precedes the process of sequentially storing each frame as a two-dimensional image.
- the image conversion apparatus 200 may divide the two-dimensional image corresponding to the two-dimensional video into a plurality of partial objects, set stereoscopic information, and convert the three-dimensional image into a three-dimensional image.
- the image conversion apparatus 200 may further perform a process of converting the generated 3D images into a 3D video.
- the image conversion apparatus 200 may set a zero point corresponding to the two-dimensional image and convert the image into a three-dimensional image using the zero point.
- the converted three-dimensional image may include both intaglio and emboss.
- the zero point is defined as stereoscopic information positioned on a display plane, not an intaglio or an embossment, in a 3D image or a 3D video. That is, partial objects corresponding to the zero point are located on the display plane. That is, each of the partial entities based on the zero point is represented by an intaglio or an embossment in correspondence with the set stereoscopic information.
- FIG. 3 is a block diagram illustrating an internal functional configuration of an image conversion apparatus according to an embodiment of the present invention.
- the image conversion apparatus 200 includes a setting unit 310, a storage unit 315, a generation unit 320, a synthesis unit 325, and a control unit 330.
- the setting unit 310 classifies the two-dimensional image from the central object into the partial objects under the control of the controller 330, and sets depth information for each partial object.
- the setting unit 310 sets contour information on regions of the partial objects having the same depth information.
- the contour information is to define that the depth information is the area of the partial entities.
- the stereoscopic information includes contour information on regions for partial objects having the same depth information and / or depth information corresponding to the contour information.
- the partial objects are defined as areas having n-th depth information at an n-th (arbitrary natural number) time point in the original image. Therefore, the partial entities separated at the n-th time point are set in an area in which the partial entities separated up to the n-th time point are excluded.
- the setting unit 310 may arrange and output stereoscopic information on partial objects having the same depth information by using a layer.
- an nth layer may be generated at an nth view, and partial objects separated at the nth view may be arranged and output using a one-dimensional array.
- the n-th layer may also include depth information of the partial objects included in the n-th layer.
- each layer has a hierarchical connection structure. That is, the depth information of the partial objects included in the n-th layer may be the depth information of the final n-th layer, which is the sum of the depth information of the n-th layer.
- depth information of each layer may be a difference value from depth information of an adjacent upper layer.
- each layer may include depth information of each partial object.
- the image conversion apparatus 200 may convert a two-dimensional image into a three-dimensional image. This is the same as the process of converting a two-dimensional video to a three-dimensional video.
- the storage unit 315 includes software for operating the image conversion apparatus 200 according to the present invention, a plurality of two-dimensional images, two-dimensional videos, converted three-dimensional images, converted three-dimensional videos, and parallax temporary images generated during the conversion process. Stored.
- the generation unit 320 performs a function of generating the first image and the second image by using stereoscopic information and / or zero point under the control of the controller 330.
- the generation unit 320 generates the first image by moving the two-dimensional image in a predetermined direction (left or right) using the stereoscopic information set by the setting unit 310.
- the generation unit 320 may generate a second image by moving the two-dimensional image in a predetermined direction by using a zero point under the control of the controller 330.
- the moving direction of the two-dimensional image for generating the second image may be opposite to the direction in which the two-dimensional image is moved to generate the first image.
- the generation unit 320 may generate the second image by moving the two-dimensional image to the right direction corresponding to the zero point.
- the generation unit 320 receives a zero point from the control unit 330.
- the generation unit 320 generates movement information by using the total depth information set for the input zero point and the two-dimensional image.
- the generation unit 320 may generate the second image by moving the two-dimensional image in a predetermined direction by using the movement information.
- the movement information is calculated by the generation unit 320, and a description thereof will be focused.
- the movement information may be calculated by the control unit and input to the generation unit 320.
- the synthesizer 325 performs a function of synthesizing the first image and the second image generated by the generator 320 to generate a three-dimensional image.
- the two-dimensional image is moved in the first direction according to the stereoscopic information to generate the first image, and the two-dimensional image is moved in the second direction according to the zero point (that is, the movement information) to generate the second image.
- the zero point that is, the movement information
- the zero point is depth information located on the display plane
- the partial information corresponding to the zero point in the generated three-dimensional image has the depth information as the default value (eg, 0) because the depth information is located on the display plane. It can be expressed as.
- the first image in which the partial objects are moved according to the depth information corresponding to each partial object and the second image in which the two-dimensional image is moved corresponding to the zero point are synthesized, and consequently, each part is engraved and embossed based on the set zero point.
- the advantage is that the objects are represented.
- the controller 330 is an internal component of the image conversion apparatus 200 according to the present invention (for example, the setting unit 310, the storage unit 315, the generation unit 320, the synthesis unit 325, etc.) To control. In addition, the controller 330 may receive and set the total depth information of the two-dimensional image from the outside.
- the controller 330 may determine whether the input original image is a two-dimensional image or a two-dimensional video, and if the two-dimensional video is a two-dimensional video, the control unit 330 may divide the two-dimensional video by each frame unit and store the sequential two-dimensional image.
- the controller 330 may generate a 3D video by synthesizing the converted 3D images corresponding to each frame of the 2D video.
- controller 330 may receive a zero point from the outside, calculate movement information on the two-dimensional image, and output the calculated information to the generation unit 320.
- the image conversion apparatus 200 may further include a correction unit (not shown) for correcting the converted 3D image or 3D video.
- the correction unit performs a function of correcting the 3D image or the 3D video under the control of the controller 330.
- FIG. 4 is a flowchart illustrating a method of converting a 2D video into a 3D video according to an embodiment of the present invention.
- a method of converting a 2D video into a 3D video will be described.
- each step described below is performed by each internal component of the image conversion apparatus 200, but will be collectively described as the image conversion apparatus 200 for the convenience of understanding and explanation.
- the image conversion apparatus 200 receives a 2D video.
- the image conversion apparatus 200 may perform a process of determining whether the input original image is a 2D video or a 2D image.
- the process since the description is made on the assumption that the 2D video is input, the corresponding process is omitted. However, after the process of determining the 2D video or the 2D image, the process may proceed to step 415.
- the image conversion apparatus 200 divides the 2D video in units of frames.
- the image conversion apparatus 200 converts and stores each divided frame into a sequential two-dimensional image.
- the image conversion apparatus 200 may classify the two-dimensional video into frame units and convert the two-dimensional video into a sequential two-dimensional image and store the same.
- the image conversion apparatus 200 converts (converts) the first two-dimensional image corresponding to the first frame into a three-dimensional image.
- the image conversion apparatus 200 classifies the first two-dimensional image into partial objects with respect to the center object, and sets depth information with respect to the partial objects.
- the image conversion apparatus 200 sets the region of the partial objects corresponding to the set depth information as the contour information.
- the image conversion apparatus 200 may set partial contour information and generate depth information on the remaining portions except for the set contour information, and set the next contour information.
- such contour information may be sequentially and hierarchically set. In this manner, the image conversion apparatus 200 may set stereoscopic information by dividing each partial object in order from the center object to the surrounding object.
- the stereoscopic information may include depth information and / or contour information.
- the contour information is defined as curve information connecting regions having the same depth information in the two-dimensional image. Therefore, the contour information may be expressed as area information corresponding to arbitrary depth information.
- a method of classifying partial objects and setting stereoscopic information in order from a central object to a peripheral object will be described in more detail with reference to FIG. 11 below.
- the image conversion apparatus 200 converts (ie, converts) the two-dimensional image after the first frame into a three-dimensional image.
- the image conversion apparatus 200 loads stereoscopic information set corresponding to the (n-1) (n is a natural number) two-dimensional image corresponding to the adjacent previous frame. do.
- the image conversion apparatus 200 extracts the changed objects by comparing the n-th two-dimensional image and the n-th two-dimensional image. Subsequently, the image conversion apparatus 200 sets stereoscopic information on the n-th two-dimensional image by modifying stereoscopic information corresponding to the changed objects.
- the image conversion apparatus 200 may generate stereoscopic information on the n-th two-dimensional image by loading stereoscopic information corresponding to the adjacent previous frame in the case of the 2D video.
- stereoscopic information corresponding to a previous frame there is an advantage in that time consumption due to stereoscopic information is set.
- the image conversion apparatus 200 determines whether all frames of the 2D video are converted into 3D images.
- the image conversion apparatus 200 may determine whether the last frame of the 2D video is converted to the 3D image and determine whether all the frames of the 2D video are converted to the 3D image.
- step 425 If the entire frame has not been converted to a three-dimensional image, go to step 425.
- the image conversion apparatus 200 performs correction and retouching on the plurality of converted three-dimensional images in step 435.
- the image conversion apparatus 200 divides the two-dimensional image into partial objects, sets depth information differently for each partial object, and moves each partial object by using the same. In this process, the individual parts are damaged. As such, the damaged parts appear as if the image is dragged during the synthesis process. Accordingly, the image conversion apparatus 200 may smoothly reconstruct the three-dimensional image by performing correction and retouch on the damaged areas.
- the image conversion apparatus 200 In operation 440, the image conversion apparatus 200 generates a 3D video by using the corrected 3D images.
- the image conversion apparatus 200 may check the stereoscopic sense of the two-dimensional image and the connectivity between three-dimensional images corresponding to adjacent frames.
- the image conversion apparatus 200 stores the 3D video.
- FIG. 5 is a flowchart illustrating a method of converting a 2D image into a 3D image according to an embodiment of the present invention
- FIGS. 6 to 9 are views illustrating images generated according to an embodiment of the present invention.
- a specific method of converting each two-dimensional image into a corresponding three-dimensional image will be described.
- specific embodiments of the step 420 or the step 425 will be described.
- each step described below is performed by each component of the image conversion apparatus 200, but will be collectively described as an image conversion apparatus for the convenience of understanding and explanation.
- the image conversion apparatus 200 sets total depth information of the 2D image.
- the operator can arbitrarily define the overall three-dimensional feeling of the two-dimensional image. That is, the image conversion apparatus 200 may set the total depth information and the number of contour lines by receiving the number of contour lines that can be arranged corresponding to the total depth information and the same depth information of the two-dimensional image from the outside.
- the image conversion apparatus 200 classifies the two-dimensional image into partial objects, and receives and sets depth information and contour lines corresponding to the partial objects.
- the image conversion apparatus 200 sets the center object and then divides the partial objects from the center object.
- the image conversion apparatus 200 sets depth information on the divided partial object.
- the image conversion apparatus 200 sets regions of partial objects having the same depth information as the same contour information. In this manner, the image conversion apparatus 200 may classify partial objects from the center object to the surrounding object and set stereoscopic information.
- Stereoscopic information set for the two-dimensional image may be set hierarchically.
- stereoscopic information is set for the first partial entity set at the first viewpoint
- stereoscopic information corresponding to the second partial entity set at the second viewpoint may include first stereoscopic information for the first partial entity. Can be.
- the depth information of the partial objects newly set in the second partial object is a value obtained by adding the depth information of the first partial object and the depth information of the second partial object. Can be.
- the depth information corresponding to each partial object may be difference information from the depth information of the partial object set at the previous time point, which may be arbitrarily set by an operator.
- the second partial entity may include contour information of the first partial entity.
- the image conversion apparatus 200 generates the first image by moving the respective partial objects in a predetermined direction (eg, in a horizontal direction) by using stereoscopic information set corresponding to the two-dimensional image.
- the image conversion apparatus 200 may generate a first image by moving each partial object in a horizontal and horizontal direction by using depth information set in each partial object of the two-dimensional image.
- the first image is illustrated in FIG. 8.
- the image conversion apparatus 200 sets a zero point for the two-dimensional image.
- the image conversion apparatus 200 may receive and set a zero point for a two-dimensional image from the outside.
- the zero point is depth information located on the display plane in representing the three-dimensional image. Therefore, when the depth information set to zero is represented as a three-dimensional image, the depth information is set to "0".
- the image conversion apparatus 200 calculates movement information by using the set zero point and total depth information.
- the image conversion apparatus 200 may generate movement information by using Equation 1 below.
- the movement information may be calculated as 40.
- the movement information is calculated by setting the zero point, and the two-dimensional image is moved in a predetermined direction using the zero point to generate a second image, and then synthesized with the first image.
- the zero point according to the present invention it is possible to represent the intaglio and emboss in the three-dimensional image without a complicated calculation procedure.
- by changing the zero point there is an advantage that can easily change the three-dimensional image according to the intaglio and embossed.
- the image conversion apparatus 200 generates a second image by moving the two-dimensional image in a predetermined direction by using the calculated movement information.
- the second image is an image in which each partial object is moved using movement information.
- FIG. 6 is a two-dimensional image (ie, an original image), and FIG. 7 is a second temporary image moved according to movement information. 6 and 7, it can be seen that FIG. 7 is moved by the movement information.
- the image conversion apparatus 200 In operation 540, the image conversion apparatus 200 generates a 3D image by synthesizing the first image and the second image.
- the three-dimensional image generated by synthesizing the first image and the second image is generated as a three-dimensional image including both intaglio and emboss.
- a three-dimensional image synthesized using the first image (eg, FIG. 8) and the second image (eg, FIG. 7) is illustrated in FIG. 9.
- the three-dimensional image illustrated in FIG. 9 may check an image in three dimensions by using red and blue glasses.
- a three-dimensional image was generated to check a three-dimensional effect using red and blue glasses, but according to an implementation method, the three-dimensional image may be generated to check the three-dimensional effect using devices other than the red and blue glasses.
- the three-dimensional image may be generated to check the three-dimensional effect using devices other than the red and blue glasses.
- FIG. 10 is a flowchart illustrating a three-dimensional image conversion method according to another embodiment of the present invention.
- a 3D image is generated corresponding to the first frame of the 2D video, and a method of converting the 3D image into a 3D image according to the subsequent frame will be described.
- each step described below is performed by each internal component of the image conversion apparatus 200, but will be collectively described as an image conversion apparatus for convenience of understanding and explanation.
- a two-dimensional image corresponding to a frame to be converted at the present time is loaded.
- the image conversion apparatus 200 loads stereoscopic information (that is, depth information and contour information) corresponding to a three-dimensional image corresponding to a frame converted at a previous time.
- stereoscopic information that is, depth information and contour information
- the image conversion apparatus 200 modifies stereoscopic information on the object changed in correspondence with the n-th two-dimensional image.
- the zero point assumes that a value set at the time of changing the two-dimensional image corresponding to the first frame is set as an initial value and used in a subsequent frame.
- the image conversion apparatus 200 modifies the loaded stereoscopic information to generate stereoscopic information corresponding to the n-th two-dimensional image.
- the image conversion apparatus changes depth information about the object.
- the image conversion apparatus 200 may change the contour line of the object.
- the image conversion apparatus 200 may set depth information on the added object and set contour lines on the object.
- the image conversion apparatus 200 may delete stereoscopic information about the object.
- 11 and 12 are diagrams illustrating screens provided by an image conversion apparatus according to an embodiment of the present invention.
- depth information and contour lines are input and set from a central object with respect to a two-dimensional image.
- partial entities corresponding to the same depth information are set to be included in the same contour line.
- each partial object having the same depth information by using the concept of the layer may be included in the same layer.
- contour information may be set hierarchically from the center object of the two-dimensional image.
- each layer may include depth information of the layer.
- Each layer may be arranged by outputting a plurality of partial objects having the same depth information in an array form.
- the stereoscopic information on each of the partial entities may be divided into a hierarchical form based on the same depth information.
- the partial objects corresponding to the first depth information are distinguished from the center object, and as shown in 1120, the partial objects are arranged in a horizontal direction, and the partial objects are connected to set contours. .
- the partial objects corresponding to the second depth information are divided in the remaining areas except for the partial objects corresponding to the first depth information, and stereoscopic information is set.
- the second depth information may be a difference value from the first depth information.
- the last set partial objects can be seen that the sum of the depth information of all previously set partial objects becomes the depth information of the corresponding partial object.
- instructions for performing a method of converting a 2D image into a 3D image may be implemented in a program or software form by the aforementioned methods.
Abstract
Description
Claims (23)
- 영상 변환 장치가 이차원 영상을 입체 영상으로 변환하는 방법에 있어서,원본 영상에 대한 깊이 정보를 입력받아 설정하는 단계;상기 원본 영상을 부분 개체들로 각각 구분하고, 각 부분 개체에 대해 입체 정보를 설정하는 단계;상기 입체 정보를 이용하여 상기 원본 영상을 이동시켜 제1 영상을 생성하는 단계;상기 원본 영상에 대한 영점을 입력받아 설정하는 단계;상기 영점을 이용하여 상기 원본 영상을 이동시켜 제2 영상을 생성하는 단계; 및상기 제1 영상 및 제2 영상을 합성하여 입체 영상을 생성하는 단계를 포함하는 영상 변환 방법.
- 제1 항에 있어서,상기 입체 정보는 각 부분 개체에 대한 깊이 정보 및 등고선 정보 중 어느 하나 이상을 포함하며,상기 등고선 정보는 깊이 정보가 동일한 부분 개체들의 영역정보인 것을 특징으로 하는 영상 변환 방법.
- 제1 항에 있어서,상기 영점은 디스플레이 평면에 상응하는 깊이 정보인 것을 특징으로 하는 영상 변환 방법.
- 제2 항에 있어서,각 부분 개체에 대해 입체 정보를 설정하는 단계는,상기 원본 영상에서 중심 물체를 설정하는 단계;상기 중심 물체부터 주변 물체순으로 복수의 부분 개체들을 구분하고, 구분된 부분 개체들에 깊이 정보를 설정하는 단계; 및깊이 정보가 동일한 부분 개체들에 대해 등고선 정보를 설정하는 단계를 포함하는 것을 특징으로 하는 영상 변환 방법.
- 제4 항에 있어서,상기 입체 정보를 설정하는 단계는,제n 시점에 부분 개체들을 구분하고, 상기 구분된 부분 개체들에 대해 깊이 정보를 설정하는 단계; 및상기 제n 시점에 설정된 깊이 정보에 상응하는 부분 개체들이 포함될 제n 레이어를 생성하는 단계를 포함하되,상기 제n 레이어는 동일한 깊이 정보를 갖는 부분 개체들이 배열되며,상기 제n 레이어는 제n-1 레이어에 상응하는 부분 개체들에 대한 입체 정보를 포함하며, 상기 n은 자연수인 것을 특징으로 하는 영상 변환 방법.
- 제1 항에 있어서,상기 제1 영상을 생성하는 단계는,상기 원본 영상을 입체 정보에 따라 정해진 수평 방향으로 이동시켜 상기 제1 영상을 생성하는 것을 특징으로 하는 영상 변환 방법.
- 제5 항에 있어서,상기 제1 영상을 생성하는 단계는,상기 제n 레이어에 포함된 부분 개체들을 설정된 깊이 정보에 상응하여 수평 방향으로 이동시키되,상기 깊이 정보는 상기 제n-1 레이어에 상응하는 깊이 정보와의 차이값인 것을 특징으로 하는 영상 변환 방법.
- 제1 항에 있어서,상기 제2 영상을 생성하는 단계는,상기 영점 및 상기 원본 영상에 대한 깊이 정보를 이용하여 이동 정보를 산출하는 단계; 및상기 원본 영상을 상기 이동정보에 따라 정해진 수평 방향으로 이동시켜 상기 제2 영상을 생성하는 단계를 포함하는 것을 특징으로 하는 영상 변환 방법.
- 제8 항에 있어서,상기 제2 영상을 생성하기 위한 원본 영상의 이동 방향은 상기 제1 영상을 생성하기 위한 원본 영상의 이동 방향과 상이한 것을 특징으로 하는 영상 변환 방법.
- 제1 항에 있어서,상기 원본 영상의 깊이 정보를 입력받아 설정하는 단계 이전에,상기 원본 영상이 이차원 동영상인 경우, 상기 이차원 동영상을 각 프레임 단위로 구분하고, 구분된 각각의 프레임을 순차적으로 정지 영상으로 저장하는 단계를 더 포함하는 영상 변환 방법.
- 제10 항에 있어서,상기 원본 영상이 제1 프레임이 아닌 경우, 제n-1 프레임에 상응하는 입체 정보를 로딩하는 단계;상기 제n-1 프레임과 제n 프레임을 비교하여 변경된 영역을 추출하는 단계; 및상기 변경된 영역에 상응하여 제n-1 프레임에 상응하는 입체 정보를 수정하는 단계를 더 포함하되,상기 n은 자연수이며, 상기 제n 프레임은 현재 시점에 변환할 대상 정지 영상에 상응하는 프레임이고, 상기 제n-1 프레임은 상기 제n 프레임과 인접된 이전 시점에 변환된 정지 영상에 상응하는 프레임인 것을 특징으로 하는 영상 변환 방법.
- 이차원 영상을 입체 영상으로 변환하는 영상 변환 장치에 있어서,원본 영상에 대한 깊이 정보 및 영점을 입력받아 설정하는 제어부;상기 원본 영상을 부분 개체들로 각각 구분하고, 각 부분 개체에 대해 입체 정보를 설정하는 설정부;상기 입체 정보를 이용하여 상기 원본 영상을 이동시켜 제1 영상을 생성하고, 상기 영점을 이용하여 상기 원본 영상을 이동시켜 제2 영상을 생성하는 생성부; 및상기 제1 영상 및 제2 영상을 합성하여 입체 영상을 생성하는 합성부를 포함하는 영상 변환 장치.
- 제12 항에 있어서,상기 입체 정보는 각 부분 개체에 대한 깊이 정보 및 등고선 정보 중 어느 하나 이상을 포함하되,상기 등고선 정보는 깊이 정보가 동일한 부분 개체들의 영역 정보인 것을 특징으로 하는 영상 변환 장치.
- 제12 항에 있어서,상기 영점은 디스플레이 평면에 상응하는 깊이 정보인 것을 특징으로 하는 영상 변환 장치.
- 제13 항에 있어서,상기 설정부는 상기 원본 영상에서 중심 물체를 설정하고, 상기 중심 물체부터 주변 물체순으로 복수의 부분 개체들을 구분하며, 상기 구분된 부분 개체들에 입체 정보를 각각 설정하고, 깊이 정보가 동일한 부분 개체들에 대해 등고선 정보를 설정하는 것을 특징으로 하는 영상 변환 장치.
- 제15 항에 있어서,상기 설정부는 제n 시점에 부분 개체들을 구분하고, 상기 구분된 부분 개체들에 대해 깊이 정보를 설정하며, 상기 제n 시점에 설정된 깊이 정보에 상응하는 부분 개체들이 포함되는 제n 레이어를 생성하되,상기 제n 레이어는 제n 시점에 실정된 부분 개체들을 배열하여 출력하고,상기 제n 레이어는 제n-1 레이어에 상응하는 부분 개체들에 대한 입체 정보를 포함하며, 상기 n은 자연수인 것을 특징으로 하는 영상 변환 장치.
- 제12항에 있어서,상기 생성부는 상기 원본 영상을 상기 입체 정보에 따라 정해진 수평 방향으로 이동시켜 상기 제1 영상을 생성하는 것을 특징으로 하는 영상 변환 장치.
- 제 16항에 있어서,상기 생성부는 상기 제n 레이어에 포함된 부분 개체들을 설정된 깊이 정보에 상응하여 수평 방향으로 이동시키되,상기 깊이 정보는 상기 제n-1 레이어에 상응하는 깊이 정보와의 차이값인 것을 특징으로 하는 영상 변환 장치.
- 제12 항에 있어서,상기 생성부는 상기 영점 및 상기 원본 영상에 대한 깊이 정보를 이용하여 이동 정보를 산출하고, 상기 원본 영상을 상기 이동 정보에 따라 정해진 수평 방향으로 이동시켜 상기 제2 영상을 생성하는 것을 특징으로 하는 영상 변환 장치.
- 제19 항에 있어서,상기 제2 영상을 생성하기 위해 이동되는 원본 영상의 방향은 상기 제1 영상을 생성하기 위해 이동되는 원본 영상의 방향과 상이한 것을 특징으로 하는 영상 변환 장치.
- 제12 항에 있어서,상기 제어부는 상기 원본 영상에 대한 깊이 정보 및 영점을 입력받아 설정하기 이전에,상기 원본 영상이 이차원 동영상인 경우, 상기 이차원 동영상을 각 프레임 단위로 구분하고, 구분된 각각의 프레임을 순차적으로 정지 영상으로 변환하여 저장하는 것을 특징으로 하는 영상 변환 장치.
- 제16 항에 있어서,상기 설정부는, 상기 원본 영상이 제1 프레임이 아닌 경우, 제n-1 프레임에 상응하는 입체 정보를 로딩하고, 상기 제n-1 프레임과 제n 프레임을 비교하여 변경된 영역을 추출하며, 상기 변경된 영역에 상응하여 제n-1 프레임에 상응하는 입체 정보를 수정하되,상기 n은 자연수이며, 상기 제n 프레임은 현재 시점에 변환할 대상 정지 영상에 상응하는 프레임이고, 상기 제n-1 프레임은 상기 제n 프레임과 인접된 이전 시점에 변환된 정지 영상에 상응하는 프레임인 것을 특징으로 하는 영상 변환 장치.
- 이차원 영상을 입체 영상으로 변환하는 방법을 수행하기 위해 디지털 처리 장치에 의해 실행될 수 있는 명령어들의 프로그램이 유형적으로 구현되어 있으며, 상기 디지털 처리 장치에 의해 판독될 수 있는 프로그램을 기록한 기록매체에 있어서,원본 영상에 대한 깊이 정보를 입력받아 설정하는 단계;상기 원본 영상을 부분 개체들로 각각 구분하고, 각 부분 개체에 대해 입체 정보를 설정하는 단계;상기 입체 정보를 이용하여 상기 원본 영상을 이동시켜 제1 영상을 생성하는 단계;상기 원본 영상에 대한 영점을 입력받아 설정하는 단계;상기 영점을 이용하여 상기 원본 영상을 이동시켜 제2 영상을 생성하는 단계; 및상기 제1 영상 및 제2 영상을 합성하여 입체 영상을 생성하는 단계를 수행하는 프로그램을 기록한 기록매체.
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009801307141A CN102113022A (zh) | 2008-06-12 | 2009-06-11 | 转换图像的方法及装置 |
JP2011513430A JP2011523323A (ja) | 2008-06-12 | 2009-06-11 | 映像変換方法及び装置 |
CA2727602A CA2727602A1 (en) | 2008-06-12 | 2009-06-11 | Image conversion method and apparatus |
EP09762678A EP2302593A4 (en) | 2008-06-12 | 2009-06-11 | APPARATUS AND METHOD FOR IMAGE CONVERSION |
US12/966,178 US20110142329A1 (en) | 2008-06-12 | 2010-12-13 | Method and device for converting image |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020080055309A KR100957129B1 (ko) | 2008-06-12 | 2008-06-12 | 영상 변환 방법 및 장치 |
KR10-2008-0055309 | 2008-06-12 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/966,178 Continuation US20110142329A1 (en) | 2008-06-12 | 2010-12-13 | Method and device for converting image |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2009151292A2 true WO2009151292A2 (ko) | 2009-12-17 |
WO2009151292A3 WO2009151292A3 (ko) | 2010-03-25 |
Family
ID=41417263
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2009/003151 WO2009151292A2 (ko) | 2008-06-12 | 2009-06-11 | 영상 변환 방법 및 장치 |
Country Status (7)
Country | Link |
---|---|
US (1) | US20110142329A1 (ko) |
EP (1) | EP2302593A4 (ko) |
JP (1) | JP2011523323A (ko) |
KR (1) | KR100957129B1 (ko) |
CN (1) | CN102113022A (ko) |
CA (1) | CA2727602A1 (ko) |
WO (1) | WO2009151292A2 (ko) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101121979B1 (ko) * | 2010-01-12 | 2012-03-09 | (주) 인디에스피 | 입체 영상 변환 방법 및 입체 영상 변환 장치 |
WO2011105814A2 (ko) * | 2010-02-23 | 2011-09-01 | 삼성전자 주식회사 | 다시점 정지 영상 서비스 제공 방법 및 그 장치, 다시점 정지 영상 서비스 수신 방법 및 그 장치 |
KR101055411B1 (ko) * | 2010-03-12 | 2011-08-09 | 이상원 | 입체 영상 생성 방법 및 그 장치 |
KR100980298B1 (ko) * | 2010-05-06 | 2010-09-06 | 엠진 (주) | 모델링된 2차원 영상의 3차원 영상 변환 방법 |
KR101121625B1 (ko) | 2011-04-28 | 2012-03-09 | 서울대학교산학협력단 | 양안 영상데이터 생성 단말 및 방법 |
KR101347837B1 (ko) * | 2011-11-17 | 2014-01-07 | 조영환 | 단일 2차원 이미지를 3차원 이미지로 변환하는 방법 |
WO2013081304A1 (ko) * | 2011-11-28 | 2013-06-06 | 에스케이플래닛 주식회사 | 2차원 영상을 3차원 영상으로 변환하는 영상 변환 장치, 방법 및 그에 대한 기록매체 |
WO2013081281A1 (ko) * | 2011-11-29 | 2013-06-06 | 에스케이플래닛 주식회사 | 2차원 영상을 3차원 영상으로 변환하는 영상 변환 장치와 방법 및 그에 대한 기록매체 |
CN103118269B (zh) * | 2013-02-04 | 2014-12-10 | 河北玛雅影视有限公司 | 一种基于图像扭曲技术的图像及视频2d 转3d 的方法 |
WO2016182502A1 (en) * | 2015-05-14 | 2016-11-17 | Medha Dharmatilleke | Multi purpose mobile device case/cover integrated with a camera system & non electrical 3d/multiple video & still frame viewer for 3d and/or 2d high quality videography, photography and selfie recording |
US9948912B2 (en) * | 2015-11-04 | 2018-04-17 | Mediatek Inc. | Method for performing depth information management in an electronic device, and associated apparatus and associated computer program product |
CN105791803B (zh) * | 2016-03-16 | 2018-05-18 | 深圳创维-Rgb电子有限公司 | 一种将二维图像转化为多视点图像的显示方法及系统 |
Family Cites Families (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3260570B2 (ja) * | 1994-11-29 | 2002-02-25 | 株式会社東芝 | 動き検出装置 |
JPH08182023A (ja) * | 1994-12-26 | 1996-07-12 | Sanyo Electric Co Ltd | 2次元画像を3次元画像に変換する装置 |
KR100414629B1 (ko) * | 1995-03-29 | 2004-05-03 | 산요덴키가부시키가이샤 | 3차원표시화상생성방법,깊이정보를이용한화상처리방법,깊이정보생성방법 |
JPH08331607A (ja) * | 1995-03-29 | 1996-12-13 | Sanyo Electric Co Ltd | 三次元表示画像生成方法 |
JPH09115002A (ja) * | 1995-10-20 | 1997-05-02 | Sony Corp | 座標変換装置および方法 |
JPH09185712A (ja) * | 1995-12-28 | 1997-07-15 | Kazunari Era | 三次元画像データ作成方法 |
EP0898245B1 (en) * | 1997-08-05 | 2004-04-14 | Canon Kabushiki Kaisha | Image processing method and apparatus |
JP2000078611A (ja) * | 1998-08-31 | 2000-03-14 | Toshiba Corp | 立体映像受信装置及び立体映像システム |
KR100321897B1 (ko) * | 1998-12-22 | 2002-05-13 | 박남은 | 입체영상이미지변환방법및그장치 |
JP4578653B2 (ja) * | 2000-08-24 | 2010-11-10 | 三菱電機株式会社 | 奥行き画像生成装置、奥行き画像生成方法およびその方法をコンピュータに実行させるプログラムを記録したコンピュータ読み取り可能な記録媒体 |
DE10113211A1 (de) * | 2001-03-18 | 2002-09-19 | Robert Massen | Markierungssystem für die automatische photogrammetrische Digitalisierung von Körpern und Körperteilen |
JP4251786B2 (ja) * | 2001-05-11 | 2009-04-08 | ソニー株式会社 | 情報処理装置及び方法、並びにプログラム |
DE10156908A1 (de) * | 2001-11-21 | 2003-05-28 | Corpus E Ag | Kostengünstige Erfassung der Raumform von Körpern |
AU2003257611A1 (en) * | 2002-08-20 | 2004-03-11 | Kazunari Era | Method and device for creating 3-dimensional view image |
JP2004145832A (ja) * | 2002-08-29 | 2004-05-20 | Sharp Corp | コンテンツ作成装置、コンテンツ編集装置、コンテンツ再生装置、コンテンツ作成方法、コンテンツ編集方法、コンテンツ再生方法、コンテンツ作成プログラム、コンテンツ編集プログラム、および携帯通信端末 |
KR100436904B1 (ko) * | 2002-09-06 | 2004-06-23 | 강호석 | 2차원이미지에 대한 입체영상생성방법 |
JP4251907B2 (ja) * | 2003-04-17 | 2009-04-08 | シャープ株式会社 | 画像データ作成装置 |
JP4283606B2 (ja) * | 2003-06-19 | 2009-06-24 | 株式会社ソフィア | 遊技機 |
JP4052331B2 (ja) * | 2003-06-20 | 2008-02-27 | 日本電信電話株式会社 | 仮想視点画像生成方法及び3次元画像表示方法並びに装置 |
US20080278487A1 (en) * | 2005-04-07 | 2008-11-13 | Nxp B.V. | Method and Device for Three-Dimensional Rendering |
EP1997070B1 (de) * | 2006-03-22 | 2018-01-03 | Pilz GmbH & Co. KG | Raum-zeitliches verfahren und vorrichtung zur korrespondenzbestimmung, vorzugsweise zur dreidimensionalen rekonstruktion einer szene |
CA2653815C (en) * | 2006-06-23 | 2016-10-04 | Imax Corporation | Methods and systems for converting 2d motion pictures for stereoscopic 3d exhibition |
-
2008
- 2008-06-12 KR KR1020080055309A patent/KR100957129B1/ko not_active IP Right Cessation
-
2009
- 2009-06-11 CA CA2727602A patent/CA2727602A1/en not_active Abandoned
- 2009-06-11 EP EP09762678A patent/EP2302593A4/en not_active Withdrawn
- 2009-06-11 CN CN2009801307141A patent/CN102113022A/zh active Pending
- 2009-06-11 WO PCT/KR2009/003151 patent/WO2009151292A2/ko active Application Filing
- 2009-06-11 JP JP2011513430A patent/JP2011523323A/ja active Pending
-
2010
- 2010-12-13 US US12/966,178 patent/US20110142329A1/en not_active Abandoned
Non-Patent Citations (1)
Title |
---|
None |
Also Published As
Publication number | Publication date |
---|---|
CN102113022A (zh) | 2011-06-29 |
KR20090129175A (ko) | 2009-12-16 |
KR100957129B1 (ko) | 2010-05-11 |
WO2009151292A3 (ko) | 2010-03-25 |
EP2302593A4 (en) | 2013-01-23 |
CA2727602A1 (en) | 2009-12-17 |
US20110142329A1 (en) | 2011-06-16 |
JP2011523323A (ja) | 2011-08-04 |
EP2302593A2 (en) | 2011-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2009151292A2 (ko) | 영상 변환 방법 및 장치 | |
WO2017010695A1 (en) | Three dimensional content generating apparatus and three dimensional content generating method thereof | |
WO2020032354A1 (en) | Method, storage medium and apparatus for converting 2d picture set to 3d model | |
WO2018090455A1 (zh) | 一种终端全景图像处理方法、装置及终端 | |
WO2020226317A1 (en) | Image processing apparatus and image processing method thereof | |
WO2013118955A1 (ko) | 깊이 맵 보정 장치 및 방법과 이를 이용한 입체 영상 변환 장치 및 방법 | |
WO2013111994A1 (en) | Image processing method and apparatus for 3d video | |
WO2018093100A1 (en) | Electronic apparatus and method for processing image thereof | |
WO2019164185A1 (en) | Electronic device and method for correcting image corrected in first image processing scheme in external electronic device in second image processing scheme | |
WO2017026705A1 (ko) | 360도 3d 입체 영상을 생성하는 전자 장치 및 이의 방법 | |
WO2018066760A1 (ko) | 다수의 카메라를 이용한 최적의 구형 영상 획득 방법 | |
WO2022255529A1 (ko) | 머신 러닝 기반의 립싱크 영상 생성을 위한 학습 방법 및 이를 수행하기 위한 립싱크 영상 생성 장치 | |
WO2022169035A1 (ko) | 영상 품질을 향상시킬 수 있는 영상 합성 장치 및 방법 | |
WO2020105788A1 (en) | Method and system for generating multi-faceted images using virtual camera | |
WO2020040521A1 (ko) | 라이트 필드의 중간 뷰 합성 방법, 라이트 필드의 중간 뷰 합성 시스템과 라이트 필드 압축 방법 | |
WO2020101434A1 (ko) | 영상의 리타겟팅을 위한 영상 처리 장치 및 방법 | |
WO2013077522A1 (en) | Apparatus and method for hierarchical stereo matching | |
WO2020122478A2 (ko) | 기계 학습 기반으로 파라미터를 학습하는 영상 처리 장치 및 동작 방법 | |
WO2022169036A1 (ko) | 영상 품질을 향상시킬 수 있는 영상 합성 장치 및 방법 | |
WO2019160237A1 (en) | Electronic device and method for controlling display of images | |
WO2019103240A1 (ko) | Ip를 통한 멀티카메라 비디오 분배 및 결합 시스템 및 그 방법 | |
WO2020190030A1 (en) | Electronic device for generating composite image and method thereof | |
WO2022080680A1 (ko) | 인공지능 기반 이미지 인페인팅 방법 및 디바이스 | |
WO2023101167A1 (ko) | 후면 영상정보 생성 장치 및 방법 | |
WO2012086912A1 (ko) | 2차원 영상을 스테레오 영상으로 변환하는 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980130714.1 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09762678 Country of ref document: EP Kind code of ref document: A2 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009762678 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011513430 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2727602 Country of ref document: CA |
|
NENP | Non-entry into the national phase |
Ref country code: DE |