KR101390961B1 - Conversion system for two dimensional image to three dimensional image, method thereof and recordable medium storing the method - Google Patents

Abstract

The present invention relates to an image conversion system and method for converting a two-dimensional image into a three-dimensional image, and a recording medium therefor. More particularly, the present invention relates to an object- Device; A depth map generation unit configured to store and manage a main object and a dependent object as a group by setting mutual dependency relation between two or more objects selected by the user input device and to recognize a depth information set for each object, ; A display device for displaying the two-dimensional image and the selected object; And an image control device displaying the two-dimensional image on the display device and displaying the dependent object when the main object of the two-dimensional image is selected. Accordingly, it is possible to efficiently manage a plurality of objects. In the case of inputting depth information of an object, an object is displayed in a group unit so that objects affected by depth information of a predetermined object can be confirmed at a glance, The convenience of inputting information can be improved and detailed editing results can be obtained.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image conversion system and method for converting a two-dimensional image into a three-dimensional image, and a recording medium therefor. 2. Description of the Related Art [0002]

The present invention relates to an image conversion system and method for converting a two-dimensional image into a three-dimensional image, and more particularly to an image conversion system and method for converting a two-dimensional image into a three- In order to select the depth information, one or more dependent objects included in the main object area are set so that the main object and the dependent objects can be managed as one group, thereby efficiently managing a plurality of objects In case of inputting the depth information of the object, it is possible to identify the objects which are influenced by the depth information of the predetermined object at a glance by displaying the object in the group unit, thereby improving the convenience of inputting depth information and obtaining detailed editing results Image conversion system and method for converting a two-dimensional image into a three-dimensional image For a recording medium.

As interest in 3D stereoscopic images increases, various methods for generating stereoscopic images are being developed. Among the methods of generating a stereoscopic image, there is a method of converting a two-dimensional video image into a three-dimensional stereoscopic image. When a two-dimensional video image is converted into a three-dimensional image, it is possible to utilize the already-created video content, and it is possible to obtain a three-dimensional image without using a separate photographing device.

In order to convert a 2D plane image into a 3D stereoscopic image, it is necessary to generate a left image and a right image having different parallax information from a 2D plane image. Accordingly, a depth map is generated by providing depth information to objects displayed on a two-dimensional plane image, and a left image and a right image are generated based on the depth map, thereby generating a three- . Thus, the quality of the 3D stereoscopic image greatly depends on the accuracy of the depth map.

On the other hand, as a method of generating a depth map, there is a method of extracting an object from a two-dimensional image and setting depth information for each object. The depth information set for each object is applied to the corresponding objects of the three-dimensional image rendering to generate an image having a three-dimensional sense as a whole. Therefore, the higher the number of objects to which the depth information is given, the higher quality three-dimensional stereoscopic image can be obtained.

However, since the image conversion system according to the related art manages the objects of the two-dimensional image independently, only the selected object is displayed when the operator sets or edits the depth information of the object. Therefore, as the number of objects increases, it becomes difficult to manage them, and thus the number of objects that can be set is limited.

In addition, when setting or editing depth information of objects located in mutually related areas, the user has difficulty in selecting and editing them individually.

Korean Patent Publication No. 2011-0099526 A method of converting a plane image into a stereoscopic image

SUMMARY OF THE INVENTION The present invention has been conceived to solve the above-described problems, and it is an object of the present invention to provide an image processing apparatus and method, in which, when a user selects an object in a two- The main object and the dependent object can be managed as one group by setting one or more dependent objects included in the object, thereby enabling efficient management of a plurality of objects. In the case of inputting the depth information of the object, Dimensional image to convert the two-dimensional image into a three-dimensional image so as to improve the convenience of inputting depth information and obtain detailed editing results by allowing the user to confirm the objects to be influenced by the depth information of a predetermined object at a glance, A system and method, and a recording medium therefor.

According to an aspect of the present invention, there is provided a method of generating a two-dimensional image, the method comprising: inputting object information defining at least two objects in the two-dimensional image and depth information of the object; A depth map generation unit configured to store and manage a main object and a dependent object as a group by setting mutual dependency relation between two or more objects selected by the user input device and to recognize a depth information set for each object, ; A display device for displaying the two-dimensional image and the selected object; And an image control device displaying the two-dimensional image on the display device and displaying the dependent object together when the main object of the two-dimensional image is selected.

Here, the depth map generating device may include an object information extracting module for storing and managing a main object and a dependent object as one group by establishing mutual dependency relation between two or more objects selected from the user input device; A depth information recognition module that recognizes a Z value input to the user input device as depth information of an object; And a depth map for generating a depth map based on the recognition information of the recognized object.

The object information extraction module may include an object information input / output unit for receiving the object information and outputting object information requested by the image control apparatus; An inter-object main dependency setting unit for classifying the input object as a main object and a subordinate object subordinate to the main object; And an object information storage unit in which the main object and the dependent object are stored.

Here, the main object is an object defined as an area including other objects among the input objects, and the dependent object may be an object included in the area of the main object.

According to another aspect of the present invention, there is provided a method for managing a two-dimensional image, the method comprising: storing and managing a main object and a dependent object in a group by setting mutual dependency relations for at least two or more object information defining the object in the two- An object information extraction module; A depth information recognition module for recognizing depth information input to the object; And a depth map generation module that generates a depth map based on the depth information input to each object.

Here, the object information extraction module may include an object information input / output unit for inputting / outputting the object information; An inter-object main dependency setting unit which sets an object defined as an area including other objects among the input objects as a main object and sets the objects included in the main object area as a dependent object; And an object information storage unit for storing the main object and the dependent objects as a group and providing the main object and the dependent object together when the output of the main object is selected through the object information input / output unit.

According to another aspect of the present invention, there is provided a method of transforming a two-dimensional image, the method comprising: displaying a two-dimensional image input for three-dimensional transformation in a three-dimensional space; Receiving object information defining at least two objects in the two-dimensional image; Storing a main object and a dependent object as a group by setting a dependency relation on the two or more objects; Displaying the dependent object together when the main object is selected; Receiving depth information of the object and generating a depth map; And rendering the two-dimensional image as a three-dimensional image according to the depth map.

In this case, the step of setting the dependency relation between the two or more objects and storing the main object and the dependent objects as one group may include setting an object defined as an area including other objects among the input objects as a main object , And setting the objects included in the area of the main object as a dependent object.

The step of receiving the depth information of the object and generating the depth map may include receiving the depth value of the object by receiving the Z value of the object.

According to another aspect of the present invention, there is provided an image processing method comprising: inputting object information defining at least two objects in a two-dimensional image; Setting a dependency relation between the two or more objects to set a main object and a dependent object, and storing the main object and the dependent object as a group; Displaying the dependent object together when the main object is selected; Receiving depth information of the main object and the dependent object; And generating a depth map based on the depth information input to each of the objects.

The step of setting a dependency relation between the two or more objects to set a main object and a dependent object and storing the main object and the dependent object as a group may further include the step of storing an object defined as an area including other objects among the input objects as a main object And setting objects included in an area of the main object as a dependent object.

According to another aspect of the present invention, there is provided a method of converting a two-dimensional image into a three-dimensional image, the method comprising: receiving object information defining at least two objects in the two-dimensional image; Storing a main object and a dependent object as a group by setting a dependency relation on the two or more objects; Displaying the dependent object together when the main object is selected; Receiving depth information of the object and generating a depth map; And rendering the two-dimensional image as a three-dimensional image according to the depth map, wherein the image is recorded by a program and readable in an electronic device.

As described above, in order to convert a two-dimensional image according to the present invention into a three-dimensional image, when a user selects an object in the two-dimensional image and inputs depth information, one or more dependent objects included in the region of the main object are set So that the main object and the dependent object can be managed as one group, thereby efficiently managing a plurality of objects.

In addition, when the depth information of the object is inputted, the objects are displayed in group units, so that the objects that are affected by the depth information of the predetermined object can be confirmed at a glance, thereby improving the convenience of inputting depth information and obtaining detailed editing results Thus, there is an effect that the quality of the converted three-dimensional image can be improved.

1 is a control block diagram of an image conversion system for converting a two-dimensional image into a three-dimensional image according to an embodiment of the present invention,
2 is a control block diagram of a depth map processing apparatus according to an embodiment of the present invention,
FIG. 3 is a control block diagram of the object information extracting unit of FIG. 2,
FIG. 4 is a flowchart of an image conversion method for converting a two-dimensional image into a three-dimensional image according to an embodiment of the present invention;
5 is a display state diagram of a work screen of the image conversion system according to the embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description with reference to the accompanying drawings, the same or corresponding components will be denoted by the same reference numerals, and redundant description thereof will be omitted.

1 is a control block diagram of an image conversion system 100 for converting a two-dimensional image into a three-dimensional image according to an embodiment of the present invention. The image conversion system 100 according to the embodiment of the present invention provides a graphical interface for the user to set depth information on the input two-dimensional image, generates a depth map according to the input of the user, Dimensional image.

1, the image conversion system 100 of the present invention includes a two-dimensional image input device 110, a user input device 120, a display device 130, an image control device 140, Device 200 and a three-dimensional image rendering device 160. [

The two-dimensional image input apparatus 110 receives a two-dimensional image to be converted into a three-dimensional image.

The user input device 120 provides a user interface for selecting and controlling functions of the image conversion system 100 and transmits the input of the user to the image control device 140. The user can select the object displayed in the two-dimensional image as a closed curve through the user input device 120 and input depth information to the object. The object selected in the image may be automatically selected using an image recognition technology such as edge recognition or may be selected by setting a Bezier curve through the user input device 120. [ Bezier curves are irregular curves generated by selecting several control points and connecting the points. In addition, the user can select the object through the user input device 120 and input depth information.

The display device 130 displays a series of work processes for inputting depth information for the input two-dimensional image and image conversion, input results, processing results, and the like to the user. Dimensional image in the x, y plane of the three-dimensional space and the object to which the depth information is input is located in the space shifted in the Z-axis direction by the inputted depth on the screen of the display device 130 for inputting the depth information. Thus, the user can intuitively recognize and edit the depth information of each object.

The depth map generating apparatus 200 generates a depth map based on depth information of an object input by the user. The object can be selected automatically from the Bezier curve by the user in the two-dimensional image, or automatically selected using the image recognition function such as edge detection technique, or the object can be selected in parallel with the automatic and manual operations. The object information input and output through the object information extraction module 210 may include identification information of the object selected in the image, x, y coordinates, and the like.

Here, in the depth map processing, the user directly selects an object through the user input device 120, and sets and edits a dependent object within the main object. The depth information setting and editing processes of the depth map generating apparatus 200 can be confirmed through the display device 130.

The three-dimensional image rendering device 160 converts the input two-dimensional image into a three-dimensional image using the depth map generated by the depth map generation device 200. The 3D image rendering device 160 generates a left image and a right image having different parallax information using a depth map and combines a left image and a right image having different parallax information to generate a 3D image .

The image control device 140 displays the two-dimensional image input to the image conversion system 100 on the display device 130 and displays the depth of the object selection signal and the depth map setting signal input by the user input device 120, To the map processing device.

The image control device 140 displays the process of setting, editing, and generating a depth map, which is processed through the depth map processing device, on the display device 130. The image control device 140 displays on the display device 130 a process of the user selecting a specific object in the two-dimensional image and moving the object in the Z-axis direction to input depth information.

The image control device 140 controls the 3D image rendering device 160 using the depth map generated through the depth map processing device to generate a 3D image and displays the 3D image on the display device 130 .

Hereinafter, the configuration of a depth map processing apparatus according to an embodiment of the present invention will be described in detail with reference to FIG. 2 and FIG. FIG. 2 is a control block diagram of a depth map processing apparatus according to an embodiment of the present invention. FIG. 3 is a control block diagram of the object information extraction module 210. FIG.

As shown in FIG. 2, the depth map processing apparatus includes an object information extraction module 210, a depth information recognition module 220, and a depth map generation module 230.

The object information extraction module 210 extracts and stores the information of the object selected for setting the depth information in the two-dimensional image and stores the stored object information when the depth information modification request of the object is input through the user input device 120 To the control device. The object information extraction module 210 sets the objects set for the depth map as a main object and a dependent object included in the main object area and stores the set object as a dependent object. When the main object information is requested, It also provides information about

As shown in FIG. 3, the object information extraction module 210 includes an object information input / output unit 310, an inter-object main dependency setting unit 312, and an object information storage unit 314.

The object information input / output unit 310 receives object information or outputs stored object information at the request of the image control apparatus 140. The object information may include x, y coordinates for identifying the object, identification information, and the like. In the two-dimensional image, the object can be selected directly by Bezier curves or can be selected automatically by using the image recognition function such as edge detection technique, and it is also possible to select the object in parallel with automatic and manual operations.

The inter-object main dependency setting unit 312 classifies the input object into a main object and a dependent object dependent on the main object, and stores the classified object in the object information storage unit 314. The inter-object main dependency setting unit 312 sets a group of dependent objects in units of the main object. Dependent objects can be set to one or more within the scope of the main object, and when the main object is selected, the group of dependent objects belonging to the main object are also automatically selected. On the other hand, the main dependency relationship among the objects may be set automatically by the inter-object main dependency setting unit 312, or it may be possible for the user to directly set the main /

The object information storage unit 314 stores a main object and a group of dependent objects included therein. That is, the main object 1 may include a dependent object 1, a dependent object 2, ... a group of dependent objects N. [ Therefore, even if the number of objects to be set for setting depth information increases, it can be easily managed mainly on the main object.

When the objects of the two-dimensional image are selected through the object information input / output unit 310, the inter-object main dependency setting unit 312 determines the main-object relations of the objects and outputs the main object information and the main object region And stores a plurality of dependent objects included in the object information storage unit 314. Then, when the main object is selected, dependent objects included in the main object are also selected and outputted through the object information input / output unit 310. [

The depth information recognition module 220 extracts depth information of the object according to the Z value set in the object. The selected object in the two-dimensional image is positioned at a predetermined Z point in the three-dimensional space according to the user's selection. The depth information recognition module 220 can recognize the depth information of the object using the Z value of the object. In the present invention, if one main object is selected, the dependent objects set in the main object area are all displayed together so that depth information of each object can be input. Accordingly, when the depth of the main object is set, the depth of the dependent object included therein can be set by editing based on the depth of the main object, so that the depth information can be easily input.

The depth map generation module 230 generates a depth map of the input two-dimensional image based on the depth information of each object recognized by the depth information recognition module 220.

According to this configuration, the user can manage the main object and the dependent objects included in the main object as a group in the two-dimensional image that is being worked on. Accordingly, a plurality of objects can be efficiently managed, and a larger number of objects can be set, thereby improving the quality of a three-dimensional image.

Since the main object and the dependent objects included therein can be selected at a time and group-based editing can be performed, input and editing of depth information can be easily performed. Also, when the depth of the main object is set, the depth of the dependent object included therein can be set by editing based on the depth of the main object, so that the depth information can be easily input.

4 is a flowchart of an image conversion method for converting a two-dimensional image into a three-dimensional image according to an embodiment of the present invention.

When a two-dimensional image is input to the image conversion system 100 (S110), the image control device 140 displays the two-dimensional image on the XY plane of the three-dimensional space.

The user can check the two-dimensional image on the display device 130 and select the first object through the user input device 120 (S112).

Also, the user can select and input the second object from the two-dimensional image (S114).

The inter-object main dependency setting unit 312 of the object information extraction module 210 extracts the information of the first object and the second object, and confirms the regions of the two objects (S114).

If the second object is included in the area of the first object, the inter-object main dependency setting unit 312 sets the first object as the main object and sets the second object as the dependent object of the first object (S116) . Here, it is also possible to set the master / slave relationship between objects directly by the user.

The objects with mutual master relations are managed on a group basis, and when the main object is selected, dependent objects included in the main object are selected together. Accordingly, when the depth information of the object is inputted, the main object and the dependent object can be interlinked and edited (S118).

As described above, according to the present invention, a main object and a dependent object are managed in a group unit, and when a main object is selected and edited, the dependent objects are automatically selected. Accordingly, when the depth information of the dependent objects is inputted, the depth information input can be completed at the level of editing the depth information of the main object which is already set, thereby facilitating the depth information input operation. In addition, by setting and managing a plurality of dependent objects within the main object, the number of objects to be managed increases, and a detailed depth map can be obtained.

The operation process of the image conversion system 100 according to the embodiment of the present invention will be described with reference to FIG.

5, in the display device 130 of the image transformation system 100 according to the embodiment of the present invention, a two-dimensional image is positioned in the x and y planes and the object is moved in the Z axis direction, You can enter information.

The user can set the depth information by selecting the face of the person in the two-dimensional image as the object A and moving it in the Z-axis direction.

Thereafter, objects a-1 and a-2 corresponding to the eyes in the object A can be selected.

Since the object a-1 and the object a-2 are selected in the object A, the object A corresponding to the face is set as the main object, and the object a-1 and the object a-2 corresponding to the eye are set as the dependent object. Accordingly, the depth information can be set by adjusting the depth information of the object a-1 and the object a-2 corresponding to the eye based on the depth information set on the object A corresponding to the face.

When the user selects object A, the objects a-1 and a-2, which are dependent objects, are also automatically displayed, so that the objects that are mutually influenced by the depth information can be checked at a glance . In addition, since the depth information of the object a-1 and the object a-2, which are dependent objects, can be added and input based on the depth information of the main object A, the depth information can be easily input and edited.

In addition, since another object such as nose, leaf, eyebrow, etc. can be managed as a dependent object of the object A in the object A corresponding to the face, it can be managed together with the face object of the object A, have.

Thus, those skilled in the art will appreciate that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the embodiments described above are to be considered in all respects only as illustrative and not restrictive. The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

In order to convert a two-dimensional image into a three-dimensional image, when a user selects an object in a two-dimensional image and inputs depth information, one or more dependent objects included in the region of the main object are set, By managing the objects in a group, it is possible to efficiently manage a plurality of objects. In the case of inputting the depth information of the objects, the objects are displayed in units of groups to be influenced by the depth information of a predetermined object A system and method for converting a two-dimensional image into a three-dimensional image, which enables a user to confirm objects at a glance, thereby improving the convenience of inputting depth information and obtaining detailed editing results, and a recording medium therefor .

100: image conversion system 110: two-dimensional image input device
120: user input device 130: display device
140: Image control device 160: 3D image rendering device
200: depth map generating device 210: object information extracting module
220: depth information recognition module 230: depth map generation module
310: Object information input / output unit 312: Main interdependency setting unit between objects
314: Object information storage unit

Claims (12)

A user input device for inputting object information defining at least two objects and depth information of the object in a two-dimensional image;
A depth map generation unit configured to store and manage a main object and a dependent object as a group by setting mutual dependency relation between two or more objects selected by the user input device and to recognize a depth information set for each object, ;
A display device for displaying the two-dimensional image and the selected object; And
Dimensional image to display the two-dimensional image on the display device, and when the main object of the two-dimensional image is selected, the dependent object is also displayed together. The method according to claim 1,
Wherein the depth map generating device comprises:
An object information extraction module for storing and managing a main object and a dependent object in one group by setting mutual dependency relation between two or more objects selected from the user input device;
A depth information recognition module that recognizes a Z value input to the user input device as depth information of an object; And
And a depth map generation module for generating a depth map based on the recognized recognition information of the object. 3. The method of claim 2,
Wherein the object information extraction module comprises:
An object information input / output unit receiving the object information and outputting object information requested by the image control apparatus;
An inter-object main dependency setting unit for classifying the input object as a main object and a subordinate object subordinate to the main object; And
And an object information storage unit for storing the main object and the dependent object. The method of claim 3,
The inter-object main dependency setting unit,
Wherein the main object is an object defined as an area including other objects among the input objects, and the dependent object is an object included in an area of the main object. The method according to claim 1,
Wherein the depth map generating device comprises:
An object information extraction module for storing and managing a main object and a dependent object as one group by setting mutual dependency relation for at least two or more object information defining the object in the two-dimensional image;
A depth information recognition module for recognizing depth information input to the object; And
And a depth map generation module that generates a depth map based on depth information input to each object. 6. The method of claim 5,
Wherein the object information extraction module comprises:
An object information input / output unit for inputting and outputting the object information;
An inter-object main dependency setting unit which sets an object defined as an area including other objects among the input objects as a main object and sets the objects included in the main object area as a dependent object; And
And an object information storage unit for storing the main object and the dependent object in one group and providing the main object and the dependent object together when the output of the main object is selected through the object information input / output unit. . Displaying a two-dimensional image input for three-dimensional conversion in a three-dimensional space;
Receiving object information defining at least two objects in the two-dimensional image;
Storing a main object and a dependent object as a group by setting a dependency relation on the two or more objects;
Displaying the dependent object together when the main object is selected;
Receiving depth information of the object and generating a depth map; And
And rendering the two-dimensional image as a three-dimensional image according to the depth map. 8. The method of claim 7,
Wherein the step of storing the main object and the dependent object as a group by setting a dependency relation for the two or more objects comprises:
Setting an object defined as an area including other objects among the input objects as a main object and setting objects included in an area of the main object as a dependent object. 8. The method of claim 7,
Wherein the step of generating the depth map by receiving the depth information of the object comprises:
And receiving the Z value of the object and recognizing the depth information of the object. Receiving object information defining at least two objects in a two-dimensional image;
Setting a dependency relation between the two or more objects to set a main object and a dependent object, and storing the main object and the dependent object as a group;
Displaying the dependent object together when the main object is selected;
Receiving depth information of the main object and the dependent object; And
And generating a depth map based on the depth information input to each object. 11. The method of claim 10,
Setting a dependency relation between the two or more objects to set a main object and a dependent object, and storing the main object and the dependent object as a group,
Setting an object defined as an area including other objects among the input objects as a main object and setting objects included in an area of the main object as a dependent object, . In converting a two-dimensional image into a three-dimensional image,
Receiving object information defining at least two objects in the two-dimensional image; Storing a main object and a dependent object as a group by setting a dependency relation on the two or more objects; Displaying the dependent object together when the main object is selected; Receiving depth information of the object and generating a depth map; And rendering the two-dimensional image as a three-dimensional image according to the depth map, wherein the image is recorded by a program and readable by an electronic device.

Images