KR101542775B1 - Apparatus and method for creating 3-dimensional contents of oriental painting - Google Patents

Abstract

The present invention relates to an apparatus and method for producing stereoscopic contents of oriental painting, and more particularly, to a stereoscopic image producing apparatus and method for extracting stereoscopic contents from a stereoscopic image, A stereoscopic information extracting unit for extracting at least one piece of stereoscopic information among a distance, a depth, a viewpoint, and a focus on an orienting image based on the analysis unit, the prior knowledge information and the analysis information of the two-dimensional image analyzing unit, And a stereoscopic image converting unit for converting the stereoscopic image into stereoscopic image content by applying image information of the content from the stereoscopic information.

Description

[0001] Apparatus and method for creating stereoscopic contents of oriental painting [0002]

The present invention relates to an apparatus and method for producing stereoscopic contents of oriental painting, and more particularly, to an apparatus and method for generating stereoscopic contents of oriental painting, and more particularly, The present invention relates to an apparatus and method for producing stereoscopic contents of Oriental painting for producing stereoscopic contents suitable for stereoscopic display.

Recently, 3DTV and stereoscopic display devices are receiving much attention, and interest in stereoscopic content production is increasing. Therefore, if cultural contents such as classical Chinese ink painting can be reconstructed in three dimensions, modern interpretation of traditional paintings and media arts can be combined, and it can be used in various fields in the aspect of edutainment.

In the past attempts, we first extracted the parts that can be animated using Korean classic paintings in terms of media art, then added the animation effects to the pictures by editing the images through the design work, and added them to the LCDs And a display job. However, this technique does not consider a stereoscopic display device, and focuses on editing, processing, and displaying input data using computer graphics technology. Therefore, there is a problem in that it requires extra manual work and time and expense of a specialist to reproduce this through stereoscopic display.

In addition, there are operations for converting a general two-dimensional image into a three-dimensional image. Based on computer vision technology, depth information is derived from perspective information and vanishing points embedded in two dimensional images, and stereoscopic effect is generated based on this information to reproduce stereoscopic display.

However, when applied to a general image, the presence and performance of the image are determined according to the characteristics of the image. The biggest disadvantage is that the depth information is inferred based on the one-point perspective method, which is a western perspective, It is difficult to apply to classical ink painting. In other words, the conventional approach to general images is not suitable for the three-dimensional transformation of classical inlaid paintings because the characteristics of the classic inlaid paintings are not shown in the general image.

It is an object of the present invention to provide an apparatus and method for producing stereoscopic contents of oriental painting, which enables stereoscopic display to be produced by converting an orientalized image into stereoscopic display based on specific information contained in oriental painting.

It is another object of the present invention to provide an apparatus and method for producing stereoscopic contents of oriental painting, which enables stereoscopic contents of oriental painting to be reproduced in a modern and stereoscopic manner using computer vision technology.

According to an aspect of the present invention, there is provided an apparatus for generating stereoscopic content of Oriental painting, comprising: a display unit for receiving stereoscopic information applied to analyze orientalization and analyzing two-dimensional information of at least one of margins, A stereoscopic image extracting unit for extracting at least one stereoscopic image from the distance, depth, viewpoint, and focus of the orientated image based on the prior knowledge and analysis information of the two-dimensional image analyzing unit; And a stereoscopic image converting unit for converting the stereoscopic image into stereoscopic image content by applying image information of the stereoscopic image from the stereoscopic image to the stereoscopic image, based on the analysis information and the stereoscopic information, .

The two-dimensional image analyzing unit extracts empty space information and light and shade information from the orientated image and extracts a composition of the orientated image from the empty space information and light and shade information .

Wherein the stereoscopic information extracting unit recognizes an object in the orienting image based on the margin information and the composition information of the orienting image and extracts the view information in the orienting image by grasping a distance relation between the objects.

And the stereoscopic information extracting unit extracts depth information of the object in the orienting image based on the density information of the orienting image.

And the stereoscopic information extracting unit extracts the focal information in the orienting image based on the shade information of the orienting image.

Wherein the prior knowledge information includes at least one of blank processing information, basic composition information, and view processing information necessary for interpreting the orientalization.

Wherein the content generation unit includes a caption inserting unit for inserting at least one caption information among music information, analysis information, and work commentary information related to the orienting image in the motion picture video, and audio data of music and commentary related to the orienting image And an audio inserting unit for inserting at least one of the audio information.

The content generation unit may further include a viewpoint adjusting unit for adjusting viewpoint information of the orienting image.

According to another aspect of the present invention, there is provided a method of generating stereoscopic content of oriental painting, comprising the steps of: inputting prior knowledge information to be used for analyzing oriental painting; receiving an empty space, light and shade, Extracting at least one piece of stereoscopic information among the distance, depth, viewpoint, and focus of the orienting image based on the prior knowledge information and the analysis information of the orienting image, Generating stereoscopic image content based on the analysis information and the stereoscopic image, and converting the stereoscopic image content to stereoscopic image content by applying image information of the content from the stereoscopic image information .

The step of analyzing the two-dimensional information may include extracting empty space information and light and shade information from the orienting image, and analyzing the composition of the orienting image from the margin information and the shade information .

Wherein the step of extracting the stereoscopic information includes a step of recognizing an object in the orienting image based on the margin information and the composition information of the orienting image, .

Extracting the stereoscopic information is characterized by extracting depth information of an object in the orienting image based on the density information of the orienting image.

The extracting of the stereoscopic information is characterized by extracting the focus information in the orienting image based on the shade information of the orienting image.

Wherein the prior knowledge information includes at least one of blank processing information, basic composition information, and view processing information necessary for interpreting the orientalization.

Wherein the step of generating the content comprises the steps of: inserting at least one subtitle information among music information, analysis information and work commentary information related to the orienting image in the moving image; And inserting audio information of at least one of the data.

The step of generating the content may further include adjusting the viewpoint information of the orienting image.

As described above, according to the present invention, stereoscopic information inherent in a two-dimensional classical color image can be extracted to generate cultural contents reproduced in a stereoscopic display. The present invention has the advantage of allowing the general public to enjoy the pictures with curiosity by revealing the value of the old painting which is felt static and difficult through the three-dimensional transformation of the ink painting which has never been attempted in the past.

The conventional two-dimensional image to three-dimensional conversion technique can be applied only to a general image and focuses on automation, so it can not be applied to an image having inherent characteristics such as ink painting. In addition, it has a disadvantage in that it is difficult to add or maximize a specific effect according to the intention of the author as it generates the stereoscopic image automatically. On the contrary, the present invention applies a single rule to all works to automatically generate low-quality stereoscopic content, but appropriately utilizes the features and prior knowledge of only ink painting to create a stereoscopic effect most suited to the work for each work It is possible to re-interpret the classical ink painting and reproduce it through stereoscopic display, thereby producing high quality stereoscopic contents maximizing the effects of media art and edutainment.

1 is a block diagram for explaining a configuration of an apparatus for generating stereoscopic contents of oriental painting according to the present invention.
2 is a block diagram for explaining a detailed configuration of a two-dimensional image analyzing unit according to the present invention.
3A to 3C are views for explaining the operation of the two-dimensional image analyzing unit according to the present invention.
4 is a block diagram for explaining a detailed configuration of a stereoscopic information extracting unit according to the present invention.
5A to 5D are views for explaining the operation of the stereoscopic information extracting unit according to the present invention.
6 is a block diagram for explaining a detailed configuration of a content generation unit according to the present invention.
FIG. 7 is a flowchart showing an operation flow for a stereoscopic content generation method of Oriental painting according to the present invention.

The present invention relates to an apparatus and method for generating a stereoscopic content using a two-dimensional orientalized image, extracting depth information and multi-viewpoint information from clues inherent in the classical Chinese ink painting, and converting stereoscopic images into three- And to provide an apparatus and method for allowing a user to make a request.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the present invention, the embodiment will be described based on the ink painting in oriental painting, but the present invention is not limited thereto.

1 is a block diagram for explaining a configuration of an apparatus for generating stereoscopic contents of oriental painting according to the present invention.

1, an apparatus for generating a stereoscopic content of oriental painting according to the present invention includes a control unit 110, a video input unit 120, a video output unit 130, A storage unit 140, a two-dimensional image analysis unit 150, a stereoscopic information extraction unit 160, a content generation unit 170, and a stereoscopic image conversion unit 180. Here, the controller 110 controls the sub-operations of the stereoscopic image contents.

The image input unit 120 is a means for receiving a two-dimensional orientated image. The image output unit 130 outputs the stereoscopic image content generated by the stereoscopic content generation apparatus to the two-dimensional stereoscopic image input through the image input unit 120.

The two-dimensional image analysis unit 150 analyzes the orientated image. At this time, the two-dimensional image analyzing unit 150 analyzes the empty space and light and shade from the orienting image. Also, the two-dimensional image analyzing unit 150 analyzes the composition of the corresponding orienting image from the margin and the shade of the orienting image, and recognizes an object included in the orienting image. The detailed configuration of the two-dimensional image analysis unit 150 will be described with reference to FIG.

Based on the analysis result of the two-dimensional image analysis unit 150 and the prior knowledge input to the orienting image, the stereoscopic information extracting unit 160 extracts the distance information, the depth information, the view information, and the focus information . The detailed configuration of the stereoscopic information extraction unit 160 will be described with reference to FIG.

Based on the stereoscopic information extracted by the stereoscopic information extraction unit 160, the content generation unit 170 partially corrects an orientalized image or inserts music, analysis information, work description, and the like as subtitles to generate content for an orientalized image do. The detailed configuration of the content generating unit 170 will be described with reference to FIG.

The stereoscopic image conversion unit 180 converts the stereoscopic image of the content into the stereoscopic image by applying stereoscopic information of the corresponding orienting image to the generated content.

In other words, the stereoscopic image conversion unit 180 converts the image information so that the content generated from the content generation unit 170 can be reproduced in the stereoscopic display device. For example, the stereoscopic image conversion unit 180 converts the edited contents into the display support format so that the multi-point display device can use the multi-point effect.

The storage unit 140 stores the two-dimensional orientated image input through the image input unit 120. The storage unit 140 stores stereoscopic image analysis results of the two-dimensional image analysis unit 150, stereo information of the orientated image extracted by the stereoscopic information extraction unit 160, The content generated as a result of the operation of the image conversion unit 180 is stored.

2 is a block diagram for explaining a detailed configuration of a two-dimensional image analyzing unit according to the present invention. 2, the two-dimensional image analyzer 150 includes a blank area analyzer 151, a density analyzer 153, and a composition analyzer 155.

The margin region analyzing unit 151 analyzes a margin region in the inputted orienting image.

In Asian paintings, the whole area including sky, water, and land is mostly marginalized. In addition, margins play an important role in recognizing objects. Therefore, the blank space can be used as important information to distinguish objects drawn in oriental painting from background areas.

The shading analyzing unit 153 analyzes shading information in the inputted orienting image.

The composition analyzing unit 155 analyzes the overall composition of the oriental painting based on the margin information in the oriental painting and the jokes information.

The information analyzed by the blank area analyzing unit 151, the shading analyzing unit 153, and the composition analyzing unit 155 is stored in the storage unit 140 and can be additionally utilized in the process of generating content later .

Of course, if prior knowledge of the oriental painting works is required, it is also possible to input related information to the two-dimensional image analysis unit 150 for analysis.

The specific operation of the blank area analyzing unit 151, the shading analyzing unit 153, and the composition analyzing unit 155 of FIG. 2 will be described with reference to FIGS. 3A to 3C.

3A shows an example of analyzing a composition from a margin in an orienting image. In Fig. 3A, the portion corresponding to the ground is treated as a margin. These margins distinguish land and people.

Accordingly, the two-dimensional image analyzing unit 150 analyzes the margins from the orienting image of FIG. 3A, and also recognizes the surrounding objects, that is, the person from the margins.

At this time, the two-dimensional image analysis unit 150 grasps the symmetry and composition of oriental painting through the margin information and the object information. Particularly, in Fig. 3A, it is understood from the margin that the corresponding oriental painting is drawn in a rectangular shape.

FIG. 3B shows an example of expressing an eye through margins in an orienting image.

As shown in FIG. 3B, the method of expressing the snow in winter using margins is called the milky method, which is one of the ink painting techniques. Unlike the method of applying white directly to the snow-covered part of the Western painting, the technique is a technique that draws another part without snow to make it look as if it were stacked on the white background of the picture.

In this way, if prior knowledge of the oriental painting is required, the related information can be further input to the two-dimensional image analyzer 150 and utilized for analyzing the oriental painting.

Therefore, the two-dimensional image analyzing unit 150 can distinguish the eyes from other objects (e.g., house, tree, etc.) in the future by extracting the margin information from the orienting of FIG. 3B.

FIG. 3C shows an example of expressing an object through a joke of an ink.

Looking at the picture in Fig. 3c, bamboo is displayed in front and behind as a joke of Ei. In other words, the bamboo in the front is expressed darkly, and the bamboo in the back is expressed more blurred.

At this time, the two-dimensional image analysis unit 150 extracts information that can distinguish the bamboo located at the front side from the bamboo located at the back side by analyzing the jokes in the orientalized image of FIG.

4 is a detailed configuration of a stereoscopic information extracting unit according to the present invention.

4, the stereoscopic information extracting unit 160 includes a depth information extracting unit 161, a viewpoint information extracting unit 163, and a focus information extracting unit 165. As shown in FIG.

The depth information extracting unit 161 extracts depth information of the orienting image based on the analysis result of the intensity analyzer 153.

The viewpoint information extracting unit 163 extracts viewpoint information of the orienting image based on the analysis result of the composition analyzing unit 155. As an example, the viewpoint information extracting unit 163 extracts the multi-viewpoint information by grasping the distance relation of objects in the work based on the prior knowledge of the orienting image.

By utilizing the multi-view information extracted by the view information extracting unit 163, the stereoscopic effect can be reproduced not only in the horizontal direction but also in the vertical direction.

The focus information extracting unit 165 analyzes the orienting image and extracts the position of the focus in the corresponding motion picture image.

The information extracted by the depth information extracting unit 161, the view information extracting unit 163 and the focus information extracting unit 165 is stored in the storage unit 140 so that the three- It is important to use.

The detailed operation of the depth information extracting unit 161, the viewpoint information extracting unit 163, and the focus information extracting unit 165 will be described with reference to FIGS. 5A to 5D.

5A is an exemplary diagram that is referred to explain the operation of extracting depth information.

Of the orientalized image analysis information from the two-dimensional image analysis unit 150, the most basic one is the shaded information, which mainly represents an object close to a dark picket line, and an object with a cloudy or light picket line behind it. Or an object that is focused by a dark picket line, and often an object that is out of focus due to a cloudy line.

Accordingly, the depth information extracting unit 161 extracts depth information of the input image based on the density information.

FIG. 5A is an image based on the depth information extracted by the depth information extracting unit 161 with respect to the image of FIG. 3C, and the distance relation is represented by a grayscale image having pixel values of 0 to 255 . Here, the closer to white 255 is, the closer to the black (0), the object is located in front.

FIG. 5B is an exemplary diagram that is referred to explain the operation of extracting viewpoint information.

As shown in Fig. 5B, the Orientalization basically moves the gaze direction from the upper right to the lower left direction. Therefore, when a distant object or arithmetic is located at the upper right corner and the nearest object is located at the lower left corner, an object located in front of the object is often displayed from the upper side to the lower side.

In the case of FIG. 5B, two oblique viewpoint movements are shown. In other words, the first point of view moves from the upper right to the lower left, and the second point moves from the left to the lower right. This point of view shows the way of turning backward from the back.

Therefore, the viewpoint information extracting unit 163 extracts two viewpoint movement information from the orienting image of FIG. 5B on the basis of the analysis result of the two-dimensional image analyzing unit 150. FIG.

FIG. 5C is an exemplary diagram that is referred to in explaining another operation of extracting viewpoint information.

The perspectives of Western painting are based on a single point of view law with one observer and no movement. In other words, it is common to have a viewer or artist outside the screen and draw what is visible at that point. As the abstraction develops, there is a case in which all the images of the object seen from various viewpoints like the Picasso of the Cubism are drawn in one picture, but the common points of the previous pictures are mostly assuming one viewpoint and one observer.

On the other hand, oriental painting has a perspective called the three-way law, which refers to the plateau method, the courtyard method, and the plains method. The plateau method is a technique for expressing the point at which the object is looked up from the bottom, the method for viewing the object at a similar height, and the method for representing the viewpoint from the high to the bottom.

In the orientalized image shown in FIG. 5C, it includes a view point of looking up an object from below, a view point of an object at a similar height, and a view point of looking downward from a high point.

That is, in FIG. 5C, P1 represents a time when an object is viewed at a similar height, and P2 represents a time when the object is viewed from the bottom up. In addition, P3 indicates a time point when looking downward from a high place.

Therefore, the viewpoint information extracting unit 163 extracts three pieces of viewpoint information from P1, P2, and P3.

FIG. 5D is an exemplary diagram that is referred to explain the operation of extracting the focus information. FIG.

As I mentioned earlier, the focus in oriental painting can be expressed as a joke of food.

In the case of FIG. 5d, the hill in the picture is focused on a dark picket line, but as it moves to the right and left, the picket line becomes blurred. Therefore, the focus information extracting unit 165 extracts the focus information from the dark fill line portion of P4 based on the shade information of the ink.

6 shows a detailed configuration of a content generation unit according to the present invention.

The content generation unit 170 generates content for an orientated image and performs editing based on the generated content on the basis of the two-dimensional image analysis information and stereoscopic information.

6, the content generation unit 170 includes an effect editing unit 171, a caption inserting unit 173, an audio inserting unit 175, and a view adjusting unit 177.

The effect editing unit 171 reflects and edits a specific effect on a part or all of the orienting image based on the stereoscopic information extracted by the stereoscopic information extracting unit 160, that is, depth information, view information, and focus information.

The caption inserting unit 173 inserts music information, analysis information, commentary of the work, and the like related to the orienting image in the movie image as captions.

The audio inserting unit 175 inserts the music associated with the orienting image and the voice data of the commentator into the generated content.

The view adjusting unit 177 adjusts the viewpoint of the corresponding orienting image by adding a multi-view point function or inserting a binocular view point function only according to the type of the stereoscopic display means for outputting the content.

Then, the stereoscopic image conversion unit 180 converts the stereoscopic effect of the content generated based on the stereoscopic information extracted from the stereoscopic information extraction unit 160 according to time, according to time, or adds animation can do.

The operation of the present invention configured as described above is as follows.

FIG. 7 shows an operation flow for a stereoscopic content generation method of Oriental painting according to the present invention.

As shown in FIG. 7, when the stereoscopic image is input (S200), the stereoscopic content generation apparatus analyzes the two-dimensional image information of the inputted stereotyped image, that is, the blank space, the joke, and the like (S210). At this time, the stereoscopic content generation apparatus analyzes the composition of the orienting image from the margin, the shade information according to the analysis result of the 'S210' process (S220).

Then, the stereoscopic content generation apparatus extracts the stereoscopic information of the inputted orienting image. At this time, the stereoscopic content generation apparatus extracts depth information, viewpoint information, and focus information of the orienting image, respectively (S230, S240, S250).

When the two-dimensional image analysis and stereoscopic information extraction process of the orienting image are completed, the stereoscopic content generation apparatus generates the content based on the analysis information and the extraction information of the steps S210 to S250 (S260).

In the 'S260' process, a subtitle insertion process, an audio insertion process, and a viewpoint adjustment process may be included.

When the content creation is completed in step 'S260', the content image is converted into a stereoscopic image in accordance with the characteristics of the stereoscopic display unit to output the content (S270).

Of course, in addition to the process of converting the content image into the stereoscopic image, the content may be converted based on the stereoscopic information extracted in the processes of S230 to S250, and the stereoscopic effect of the generated content may be converted You can also make them appear differently or add animations that vary with your gaze.

Finally, the stereoscopic image content generated in the process of 'S270' is outputted through the stereoscopic display means or the like (S280).

Although the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, . ≪ / RTI >

110: control unit 120: video input unit
130: Video output unit 140:
150: two-dimensional image analysis unit 151: margin region analysis unit
153: Tint analysis section 155: composition analysis section
160: stereoscopic information extraction unit 161: depth information extraction unit
163: viewpoint information extracting unit 165: focus information extracting unit
170: Content Generation Unit 171: Effect Editing Unit
173: Caption inserting unit 175: Audio inserting unit
177: viewpoint control unit 180: stereoscopic image conversion unit

Claims (16)

A two-dimensional image analyzing unit for analyzing two-dimensional information of at least one of margins, shades, and compositions of an orientalized image by receiving prior knowledge information applied to analyze orientalization;
A stereoscopic information extracting unit for extracting at least one piece of stereoscopic information among the distance, depth, viewpoint, and focus of the orienting image based on the prior knowledge information and analysis information of the two-dimensional image analyzing unit;
A content generating unit for generating a content for the orienting image based on the analysis information and the stereoscopic information; And
And a stereoscopic image converting unit for converting the stereoscopic image information into stereoscopic image content by applying image information of the content from the stereoscopic information,
Wherein the stereoscopic information extracting unit recognizes objects in the orienting image based on the margin information and the composition information of the orienting image, extracts view information in the orienting image by grasping a distance relation between the objects,
Wherein the stereoscopic information extracting unit extracts depth information of an object in the orienting image based on the density information of the orienting image,
Wherein the stereoscopic information extraction unit extracts the focal information in the orientated image based on the shade information of the orientated image. The method according to claim 1,
Wherein the two-
Extracting empty space information and light and shade information from the orienting image and analyzing the composition of the orienting image from the empty space information and light and shade information, Stereoscopic Content Generator for Oriental Painting. delete delete delete The method according to claim 1,
The pre-
Wherein the stereoscopic image includes at least one of white space processing information, basic composition information, and view processing information necessary for interpreting the orientalization. The method according to claim 1,
The content generation unit may generate,
A caption inserting unit for inserting at least one caption information among music information, analysis information, and work commentary information related to the orienting image in the orienting image;
And an audio inserter for inserting audio information of at least one of music and commentary voice data related to the orienting image. The method of claim 7,
The content generation unit may generate,
The stereoscopic content generating apparatus of claim 1, further comprising: a viewpoint adjusting unit for adjusting viewpoint information of the orienting image. Analyzing two-dimensional information on at least one of empty space, light and shade, and composition of an orientalized image by inputting prior knowledge information applied to analyze orientalization;
Extracting at least one piece of stereoscopic information from the distance, depth, viewpoint, and focus of the orienting image based on the prior knowledge information and analysis information of the orienting image;
Generating content for the orienting image based on the analysis information and the stereoscopic information; And
Converting the stereoscopic image into stereoscopic image content by applying image information of the content from the stereoscopic information,
Wherein the extracting of stereoscopic information includes extracting viewpoint information in the orienting image by recognizing objects in the orienting image based on margin information and composition information of the orienting image and grasping a distance relation between the objects, Extracting depth information of an object in the orienting image based on the density information of the orienting image, and extracting the focus information in the orienting image based on the density information of the orienting image. The method of claim 9,
Wherein the step of analyzing the two-
Extracting empty space information and light and shade information from the orientated image,
And analyzing a composition of the orienting image from the margin information and the density information. delete delete delete The method of claim 9,
The pre-
Wherein the stereoscopic image includes at least one of white space processing information, basic composition information, and viewpoint processing information necessary for interpreting the orientalization. The method of claim 9,
Wherein the generating the content comprises:
Inserting at least one subtitle information among music information, analysis information, and work commentary information related to the orienting image in the orienting image; And
And inserting audio information of at least one of music and commentary voice data related to the orienting image. 16. The method of claim 15,
Wherein the generating the content comprises:
And adjusting the viewpoint information of the orienting image based on the viewpoint information.

Images