KR102317873B1 - A System Providing Fast Video Rendering Service - Google Patents

Abstract

According to the present invention, provided is a system for providing a high-speed rendering service for an image, which includes: a condition setting module receiving the setting of rendering conditions including the number of frames from a user; a file input module receiving the upload of a rendering target image from the user; a rendering performance module rendering the rendering target image according to the rendering condition to generate a rendered image; and a result providing module providing the rendered image to the user.

Description

{A System Providing Fast Video Rendering Service}

The present invention performs the rendering of 3D images, which requires a large number of high-performance computers and time, through a number of computer servers at high speed and low cost, thereby reducing the burden on rendering and improving the efficiency of ordinary individuals. It relates to a system for providing a high-speed rendering service.

The conventional 3D image production process includes texture modeling for 3D graphic objects, creating and planting a character's skeleton in 3D computer animation, or rigging to make the character move by assigning a skeleton to the animation production, two-dimensional image. It takes into account external information such as light source, position, and color, and injects realism into it, and it consists of rendering to create a 3D image, and editing processes such as effects and dubbing. This conventional 3D image production process can also be confirmed through FIG. 1 .

Among them, rendering refers to a process or a technique for creating an image from a 3D scene including objects using a computer program. Rendering a 3D scene is based on light output from one or more virtual light sources, which are virtual light sources in 3D computer graphics, with lighting effects and shading effects on the 3D scene. It includes the process of applying a visual effect, such as.

Colors of 3D models included in a 3D scene may be determined by applying a lighting effect, and shadows, reflected light, transparent effects, etc. due to occlusion may be determined by applying a shading effect. These effects include the characteristics of the light emitted from the virtual light source (eg color and direction, etc.), the characteristics of the 3D models (eg color, surface normal, texture, etc.), and the relationship between the virtual light source and the 3D models. based on placement relationships, etc.

More specifically, 3D rendering is a field of computer graphics that renders 3D models. 3D rendering is used in various application areas such as 3D games, virtual reality, animation, and movies.

3D rendering techniques include ray tracing rendering, which renders a 3D model by tracing the path where light emitted from a light source is reflected from the surface of an object, direct lighting effects, as well as direct lighting reflected light reflected by an object or diffuse reflection phenomenon. There are radiosity rendering that reflects even the indirect lighting effect caused by light, etc., and rasterization rendering that renders a 3D model by converting vector data into a pixel pattern image.

Conventionally, as the demand for computer graphics rapidly increases, rendering technology for realistic image expression is rapidly changing. In particular, in order to produce a digital image, it takes a long time as well as a large number of high-performance computers to render each frame constituting an image and make it into an image.

As a prior art related to such image rendering, '3D rendering method and apparatus' is disclosed in Korean Patent Application Laid-Open No. 10-2021-0030147.

The present invention relates to a 3D rendering method and apparatus, wherein the 3D rendering method includes extracting sample points from a 3D scene to be rendered, rendering the extracted sample points, and rendering result information on the sample points and generating a rendering result image corresponding to the entire rendering based on rendering result information on the sample points and feature information of the 3D scene.

However, as it is difficult for an individual to perform image rendering as described above, a technology that can perform supply according to demand, especially high-quality rendering in a short time, compared to the increase in image rendering demand due to the development and dissemination of 3D software. There are still many shortcomings in

Therefore, in order to solve the above-described problems, there is a need to develop a rendering service capable of increasing rendering efficiency by rendering a 3D image in terms of high performance and a reasonable cost range.

A main object of the present invention is to provide a high-speed, low-cost rendering service providing system for 3D images.

Another object of the present invention is to enable correction of a rendered image.

Another object of the present invention is to enable the user to select whether to render by performing test rendering before actual rendering.

It is a further object of the present invention to enable differential settlement of costs according to rendering.

An additional object of the present invention is to allow an editing effect to be added to an image according to a user's tendency.

In order to achieve the above object, a system for providing a high-speed rendering service for an image according to the present invention includes: a condition setting module configured to receive a rendering condition including the number of frames from a user; a file input module for uploading a rendering target image from a user; a rendering execution module that renders the rendering target image according to the rendering condition to generate a rendered rendered image; and a result providing module that provides the rendered image to the user.

Furthermore, the system includes a correction request unit receiving a re-rendering request for the rendered image from the user, and a re-rendering unit for re-rendering the rendered image according to the re-rendering request to generate a corrected image; and a result correction module including a file providing unit that provides the corrected image to the user.

In addition, the system includes a test file input unit receiving a test rendering image from the user, a test rendering unit performing rendering on the test rendering image to generate a test execution image, and providing the test execution image to the user and a test rendering module including a test result providing unit, wherein the file input module receives the rendering target image from the user provided with the test execution image.

In addition, the system includes a payment module for receiving payment for the cost for the rendering from the user, wherein the rendering performing module renders the rendering target image as the payment of the cost is completed .

A system for providing a high-speed rendering service for an image according to the present invention,

1) By performing high-speed rendering through multiple server computers, it enables efficient rendering in a short time and enables high-quality video production at the same time.

2) Even if an error in the rendering target file occurs or the user does not like the work, it is possible to modify and re-render the rendered file,

3) Test rendering is enabled before actual rendering, so that users can decide whether to render or not after receiving the test rendering file.

4) Increased the rationality of payment for rendering,

5) It provides the effect of further enhancing the beauty of the image by receiving the user's preferred sentence about the video editing technique and analyzing it to give various effects.

1 is a flowchart illustrating a process of producing a 3D image;
2 is a conceptual diagram showing a schematic configuration of the system of the present invention.
Fig. 3 is a block diagram showing the overall configuration of the system of the present invention;
4 is a process diagram for a rendering system of the present invention;

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The accompanying drawings are not drawn to scale, and like reference numbers in each drawing refer to like elements.

2 is a conceptual diagram showing a schematic configuration of the system of the present invention.

Referring to FIG. 2 , the system for providing a high-speed rendering service for an image of the present invention preferably includes a user 1 and a central control server 2 .

The user 1 is a person who wants to perform rendering through the high-speed rendering service providing system of the present invention. In general, rendering requires a large number of high-performance computers and time, so it takes a lot of time for individual users (1) to perform rendering on their own, and at the same time, there are technical limitations. And you face the limits of your computer's performance. Accordingly, the user 1 can perform high-quality rendering quickly and at a low cost by performing rendering on behalf of the rendering service providing system of the present invention.

Accordingly, the user 1 uploads a 3D image file stored in his/her computer, laptop, tablet PC, etc. on the web page on which the system of the present invention is implemented, and then receives the rendered file through the system.

In the present invention, the central control server (2) is a management subject of the rendering service providing system and at the same time a subject that actually performs rendering on the image uploaded by the user 1, and plays a management role for the server PC for rendering. It can be said that it is a congruence.

The central control server 2 means hardware having a CPU and storage means for processing information, in other words, the central control server 2 includes a central processing unit (CPU) and storage means such as memory and hard disk. A program that can be executed in the central processing unit, i.e., software, is installed on the hardware base equipped with wireless communication equipment such as wired and Bluetooth, and this software can be executed. It will be described later as a constituent unit called 'wealth'.

At this time, the central control server 2 temporarily and/or permanently stores the signal (or data) processed therein. RAM (Random Access Memory, not shown) and ROM (Read-Only Memory, ROM, not shown), and may include a processor. In addition, the central control server 2 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, RAM, and ROM.

The processor may include one or more cores (not shown), a graphic processing unit (not shown), and/or a connection path (eg, a bus, etc.) for transmitting and receiving signals to and from other components.

The memory may store programs (one or more instructions) for executing and controlling a module or a unit to be described later. Programs stored in the memory may be divided into a plurality of modules according to functions.

The steps of a method or algorithm described in relation to an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. Software modules include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, Alternatively, it may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

That is, the components of the present invention may be implemented as a program (or an application) to be executed in combination with a computer, which is hardware, and stored in a medium. Components of the present invention may be implemented as software programming or software components, and similarly, embodiments may include various algorithms implemented as data structures, processes, routines, or combinations of other programming constructs, including C, C++ , Java, assembler, etc. may be implemented in a programming or scripting language. Functional aspects may be implemented in an algorithm running on one or more processors.

The configuration of these 'modules' or 'parts' or 'interfaces' is installed and stored in the storage means of the central control server 2, software executed via the CPU and memory, or a configuration of hardware such as FPGA or ASIC. it means. In this case, the configuration of 'module', 'unit', and 'interface' is not limited to hardware, and may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. As an example, 'module' or 'part' or 'interface' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, and properties. , procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables.

The functions provided by these 'modules' or 'units' or 'interfaces' may be combined into a smaller number of components and 'units' or 'modules' or additional components and 'units' or 'modules'. can be further separated.

Hereinafter, the detailed configuration and function of the central control server 2 will be described based on the macroscopic configuration.

3 is a block diagram showing the overall configuration of the system of the present invention.

Referring to FIG. 3 , the system for providing a high-speed rendering service for an image of the present invention includes a condition setting module 100 , a file input module 200 , a rendering execution module 300 , and a result providing module 400 . characterized in that

The condition setting module 100 performs a function of receiving a rendering condition for rendering an image, and the user 1 sets the rendering condition including the number of frames of the image to be rendered by the condition setting module 100. is set through

In addition, not only the number of frames in the video, but also the starting frame and the last frame of rendering, the rendering image nomenclature, image size and resolution, Arnold sample count, Maya version, rendering image naming, and the name of the rendering camera. It can be set as a rendering condition. Of course, such rendering conditions may be inputted through the user 1 , or the type of camera, image naming method, resolution, and the like may be selected by the user 1 .

The file input module 200 performs a function of receiving a rendering target image, which is a 3D image desired to be rendered, from the user 1 . To this end, the rendering service providing system of the present invention is preferably implemented in the form of a website, and the user 1 uploads his/her own rendering target image file to the rendering service providing system website and inputs the file. handle

After that, since rendering is a task that takes time and equipment, basically, the system of the present invention can receive payment for rendering. For this purpose, a payment module 700 may be further included, but this will be described later.

In addition, when the upload of the file is completed, the rendering performing module 300 included in the rendering service providing system of the present invention renders the rendering target image to generate a rendered rendered image. That is, when a rendering target image is rendered, a rendered image is generated.

At this time, preferably, in the rendering, it is based on that rendering is performed according to the rendering conditions set through the above-described condition setting module 100, and the rendering performing module of the present invention performs rendering through at least five rendering server computers. It is characterized in that it performs rendering quickly and at a reasonable cost. In addition, it is preferable to use the Arnold Renderer to perform high-quality rendering.

In such a rendering, multi-sampling may be performed on pixels subjected to pixel shading as in conventional rendering. Multi-sampling means sampling data at a plurality of positions included in one pixel-shaded pixel. For example, a plurality of data (sample values) for one pixel may be obtained by performing a depth test using depth value information, color blending, visibility check, etc. at a plurality of positions included in a pixel.

In this case, the number of multi-samples (the number of samples that are multi-sampled and included in one pixel) is included in the resolution information received from an external device (eg, a device driver) or based on the resolution information received from the external device. , may be a calculated value. Therefore, it goes without saying that rendering can be performed based on these values.

Therefore, in the rendering process, sampling of the learning-based sample points for rendering and reconstruction of the entire rendering result image based on the rendering results of the sample points may be performed. In an embodiment, rendering is performed only on some sample points considered to be major points of the 3D scene in the 3D scene, and a rendered image corresponding to the entire rendering may be generated based on the rendering results of the sample points. have.

Finally, the rendered image on which the rendering is completed may be provided to the user 1 through the result providing module 400 . Preferably, the user 1 provides the rendered image in such a way that, when the rendering is completed, the rendered image is transmitted via e-mail, or the rendered image is downloaded by accessing the site where the rendering service providing system of the present invention is implemented. can receive

Therefore, the high-speed rendering service providing system of the present invention can reduce the rendering time by 60 to 80% by rendering the image of the user 1 at a high speed at a low price, and can provide a high-quality rendered image. It has the effect that it can.

Furthermore, the rendering service providing system of the present invention may provide a correction function for a rendered image, and for this purpose, a result correction module 500 may be further included. The result correction module 500 basically includes a correction request unit 510 , a re-rendering unit 520 , and a file providing unit 530 .

The correction request unit 510 performs a function of receiving a re-rendering request from the user 1 provided with the rendered image. When the user 1 does not like the rendered image, there is an error in the file, or if the rendering condition needs to be changed, the user 1 can request re-rendering of the rendered image by inputting a re-rendering request. will do

Here, when it is necessary to change the rendering conditions, it will be basic that the changed rendering conditions should be input from the user 1 along with the re-rendering request, and further, if there is an error in the rendered image file itself, whether an error occurs or not from the user 1 can be input.

The re-rendering unit 520 performs a function of re-rendering, ie, re-rendering, a rendered image in response to a re-rendering request input from the user 1 to generate a corrected image. Here, in the case of re-rendering, a cost may occur. Of course, when re-rendering according to a condition change according to the request of the user 1, an additional cost must be paid from the user 1, of course. However, if there is an error in the rendered image file itself, it is based on not charging a separate fee for re-rendering.

The re-rendering unit 520 re-renders, that is, re-renders, the rendered image according to the rendering condition as in the configuration of the above-described rendering performing module 300 to generate a corrected image, and the rendering condition is the above-described modification request It is said that it can be changed by the configuration of the part 510 .

Finally, the file providing unit 530 performs a function of providing the re-rendered corrected image to the user 1 . This can be said to be a function similar to the result providing module 400, and in the case of a modified image, it is sent through the user 1's e-mail or uploaded on a web page in which the system is implemented so that the user 1 can view the corrected image. You can make it available for download.

Through the configuration of the result correction module 500 as described above, the user 1 can perform repeated re-rendering when the user 1 is not satisfied with the rendering result, and it is also possible to change the rendering condition every time re-rendering.

Furthermore, the rendering service providing system of the present invention may further add a function to allow the user 1 to experience test rendering before rendering the actual rendering target image. For this purpose, the system includes a test rendering module ( 600) may be further included. The test rendering module 600 basically includes a test file input unit 610 , a test rendering unit 620 , and a test result providing unit 630 .

The test file input unit 610 performs a function of receiving a test rendering image, which is an image to be test rendered, from the user 1 . In this case, the test rendering image is generally shorter in length and smaller in capacity than the above-described rendering target image, and may be said to be an image for a rendering test. In addition, when uploading the test rendering image, the user 1 may input simple conditions for the test rendering, that is, the number of frames, the starting frame of the rendering, the last rendering frame, and the like.

The test rendering unit 620 renders the test rendering image uploaded from the user 1 , that is, performs test rendering to generate a test execution image as a result. This is to render the test file so that the user 1 can experience the rendering of the system, paying attention to the fact that rendering of the actual rendering target image may take a long time and cost.

Accordingly, when the test rendering is completed, the test execution image generated through the test result providing unit 630 is provided to the user 1 . In this case, the test execution may also be provided through e-mail, or the user 1 may access the website on which the rendering system of the present invention is implemented and download the test execution image.

When performing such a test rendering, in the configuration of the file input module 200, the user 1 may upload a rendering target image after receiving the test execution image first, so that the test rendering is tested before actual rendering, that is, this After some prior experience of rendering according to the invention, it is possible to decide whether to render or not.

In addition, the system of the present invention can receive payment for rendering, for this purpose, a payment module 700 may be further included. The payment module 700 serves to receive payment for rendering from the user 1 , and in principle, payment is made before rendering. In addition, in the case of re-rendering compared to the initial rendering, a discounted cost may be generally applied compared to the initial rendering. Furthermore, in the configuration of the payment module 700 as described above, it is based on the rendering of the rendering target image as the payment of the cost is completed.

4 is a process diagram for a rendering system of the present invention.

Therefore, finally, when the rendering process of the present invention is described with reference to FIG. 4 , the user 1 accesses the web page in which the rendering service of the present invention is implemented, accesses the rendering service, performs test rendering, and test rendering After checking the result, you can input the actual rendering conditions and upload the target image. After that, payment is made, the rendering of the image is performed, and the rendering can be completed after confirming and correcting the result.

The rendering service providing system of the present invention as described above can reduce rendering costs and perform high-speed rendering, thereby providing only rendering services that are not easily performed by ordinary people.

Furthermore, the rendering service providing system of the present invention may further provide a visual effect to the rendered image. For this purpose, the system of the present invention may further include an image editing module 800 .

The image editing module 800 performs a function of providing a visual effect by applying an image editing technique to a rendered image that has been rendered. Accordingly, a more realistic and colorful 3D image can be produced, and the result providing module can provide the user 1 with a rendered image to which a visual effect is given through the image editing module.

Herein, the configuration of the image editing module 800 will be described in more detail. In general, the image editing module may further include a database for storing an editing technique by analyzing a conventional image, and an editing execution unit.

The database can be said to be the result of analyzing the type and sequence of the existing video editing technique, which is created and stored as the editing technique type data and the editing technique sequence data. Editing of the rendered image may be actually performed according to the sequence data. Here, the editing technique may include zooming, tilting, dissolve, fade, and the like.

Accordingly, more various visual effects and splendid image beauty may be provided by further providing such visual effects. In this case, in general, each user 1 may have a preference for an image editing technique.

Accordingly, in order to reflect this preference for each user 1, the system of the present invention may further include a tendency analysis module 900 . The tendency analysis module 900 receives a preference sentence for an image editing technique from the user 1 , and analyzes and processes the input preference sentence to determine user tendency.

Therefore, when the user tendency is identified in this way, the image editing technique is differentially applied to the rendered image according to the user tendency to give a visual effect. You will be able to create gorgeous images.

Here, there is no limitation on the method of identifying user tendencies, but preferably, a preferred sentence for the video editing technique is input, and the keyword is extracted from the input preferred sentence. In this way, user preferences can be identified.

Here, there is no limitation on the length of the preferred sentence, but in general, in order to reveal the emotion or preference of the user 1 in detail, the preferred sentence with rich content is more preferable. Preferred sentences may be input without limitation.

To this end, the propensity analysis module 900 includes a term extractor 910 , a frequency matrix generator 920 , a transpose matrix generator 930 , a simultaneous matrix calculator 940 , a singular value calculator 950 , and a final It may include a numerical calculation unit 960 and a main word extraction unit 970 .

The term extraction unit 910 segments the preferred sentence into morpheme units and extracts terms including adjectives and verbs, that is, a function of segmenting the preferred sentence into spaces to generate a node, and then extracting terms from each node. carry out

For example, the preferred sentence input by the user (1) is 'I hope it draws the viewer's attention. In particular, I want to leave a strong impression and draw the viewer's attention through the effect. Twelve nodes such as 'I want to make/impress/and/attract/attention/attract viewers/' are created.

Thereafter, terms including adjectives and verbs included in each node are extracted, because preference, that is, expressing emotion, is generally an adjective, and it is preferable that an action according to it is expressed in a verb form. Therefore, the term can be ‘strongly (han), show, lead, wish, leave, do’ (6 in total).

The frequency matrix generating unit 920 performs a function of generating a frequency matrix based on the frequency of the above-mentioned terms appearing in a preferred sentence. Here, the frequency matrix is generated based on the frequency of occurrence of each of the above-mentioned terms in each preferred sentence, and preferably, the frequency matrix may be generated based on Equation 1 below.

Equation 1,

는 빈도행렬,

는

번째 용어(

)가 시스템에 입력된

번째 선호 문장(

)에 출현한 빈도수,

은 생성된 용어의 개수,

은 시스템 상에 입력된 선호 문장의 개수를 나타낸다.)(here,

is the frequency matrix,

Is

second term (

) entered into the system

second preferred sentence (

), the number of occurrences in

is the number of generated terms,

represents the number of preferred sentences entered into the system.)

Here, the number of terms generated means the number of terms generated in the corresponding preferred sentence. And, for example, taking the examples of 'strongly' and 'show' among the above-mentioned terms, suppose that there are two preferred sentences inputted into the system. Here, the term 'strongly' appears twice in the first preference sentence, the term 'show' appears 3 times, and the term 'strongly' appears once in the second preference sentence, and shows' If the term appears 0 times,

A frequency matrix can be created in the form of

The transpose matrix generating unit 930 performs a function of generating a transposed matrix by transposing the above-described frequency matrix. Here, the transpose matrix is to change the row and column of the frequency matrix, and this transpose matrix can be generated in the form of Equation 2 below.

Equation 2,

는 전치행렬,

는 전치행렬의

행

열의 값,

은 생성된 용어의 개수,

은 시스템 상에 입력된 선호 문장의 개수를 의미하며,

는

번째 용어,

는

번째 선호 문장이다.)(here,

is the transpose matrix,

is the transpose matrix

line

column value,

is the number of generated terms,

is the number of preferred sentences input into the system,

Is

second term,

Is

This is the second preferred sentence.)

Here, in order to generate the transpose matrix, the above-described frequency matrix must be generated, and the transpose matrix is a form in which the positions of rows and columns are precisely swapped in the frequency matrix.

That is, as described above, the frequency matrix Q is

의 형태로 형성된다.If created as , then the transpose matrix W is

is formed in the form of

That is, the rows and columns are the same, that is, the values of row 1, column 2, row 2, column 3, and column 3 are the same, but the remaining values are created in the form of swapping rows and columns.

The simultaneous matrix calculation unit 940 performs a function of generating a term simultaneous matrix and a sentence simultaneous matrix by multiplying two matrices based on the calculated frequency matrix and the transpose matrix. Here, the term concurrent matrix may be generated by multiplying a frequency matrix by a transpose matrix, and a sentence concurrent matrix is generated by multiplying a transpose matrix by a frequency matrix. Accordingly, the term simultaneous matrix has as many rows and columns as the number of generated terms, and the sentence simultaneous matrix becomes a matrix having as many rows and columns as the number of generated preferred sentences.

First, the term simultaneous matrix is generated through Equation 3 below.

Equation 3,

는 용어 동시행렬,

는 빈도행렬,

는 전치행렬,

는 용어 동시행렬의

행

열의 값,

은 생성된 용어의 개수)(here,

is the term concurrent matrix,

is the frequency matrix,

is the transpose matrix,

is the term concurrent matrix

line

column value,

is the number of generated terms)

,

라고 가정했을 때,In the above example, the frequency matrix and the transpose matrix are each

,

Assuming that

is calculated as

The diagonal matrix value in the term simultaneous matrix generated in this way indicates how many times a specific term appears on the entire preferred sentence input into the system, and other values indicate how many times the above-mentioned two terms appear simultaneously. .

Next, the sentence simultaneous matrix can be calculated through Equation 4 below.

Equation 4,

는 문장 동시행렬,

는 빈도행렬,

는 전치행렬,

는 문장 동시행렬의

행

열의 값,

은 시스템 상에 입력된 선호 문장의 개수)(here,

is the sentence simultaneous matrix,

is the frequency matrix,

is the transpose matrix,

is the sentence concurrent matrix

line

column value,

is the number of preferred sentences entered into the system)

,

라고 가정했을 때,In the above example, the frequency matrix and the transpose matrix are each

,

Assuming that

is calculated as

The sentence simultaneous matrix is a value obtained by multiplying the transpose matrix by the frequency matrix, and has as many rows and columns as the number of preferred sentences input to the system. It represents the value of the total frequency shown.

The singular value calculating unit 950 performs a function of calculating a term singular value and a sentence singular value from each of the generated simultaneous term and sentence simultaneous matrix. Here, the singular value may preferably be an eigenvalue of a corresponding matrix.

Accordingly, the term singular value may be calculated through Equation 5 below, and the sentence singular value may be calculated through Equation 6 below.

Equation 5,

Equation 6,

는 용어 특이값,

는 문장 특이값,

는 용어 동시행렬의 고유값 중 최대값,

는 문장 동시행렬의 고유값 중 최대값,

는 용어 동시행렬,

는 문장 동시행렬)(here,

is the term singular value,

is the sentence singular value,

is the maximum value among the eigenvalues of the term simultaneous matrix,

is the maximum value among the eigenvalues of the sentence simultaneous matrix,

is the term concurrent matrix,

is a sentence simultaneous matrix)

Therefore, in order to calculate the singular value, the eigenvalue of each of the term simultaneous matrix and the sentence simultaneous matrix must be calculated. This can be calculated through conventional mathematical knowledge.

이고, 문장 동시행렬

일 때,term simultaneous matrix

, and the sentence simultaneous matrix

when,

If we compute the eigenvalues of the term concurrent matrix,

의 해를 구하면 되므로,

Since we need to find the solution of

Therefore, the maximum value among the eigenvalues of the term simultaneous matrix is 13.16,

The term singular value will be 3.63, the square root of that value.

If we calculate the eigenvalues of the sentence simultaneous matrix,

의 해를 구하면 되므로,

Since we need to find the solution of

Therefore, the maximum value among the eigenvalues of the sentence simultaneous matrix is 13.32,

The sentence singular value is 3.65, the square root of that value.

The dimensions of the generated terms are compressed based on the singular value calculated in this way, and based on the compressed dimensions, a main word, which can be called a subject, is extracted from among a plurality of terms.

To this end, the final numerical value calculation unit 960 calculates the final numerical value based on the term singular value and the sentence singular value calculated for extracting the key word. At this time, the final numerical value is calculated based on the sum of the term singular value and the sentence singular value. Alternatively, the final value may be calculated based on the average value of the term singular value and the sentence singular value.

Therefore, there is no limitation on the method of calculating the final value, and in general, the final value is calculated by adding the term singular value and the sentence singular value in calculating the final value.

Finally, the main word extraction unit 970 extracts at least one main word from the plurality of terms according to the calculated final value. In this case, in extracting the main words from the terms, preferably, the higher the final number, the higher the number of extracted main words, and the lower the final number, the lower the number of extracted main words.

Therefore, the number of key words extracted from the term is determined according to the calculated final value, and accordingly, at the level of the system administrator. Alternatively, the system can automatically extract the subject word from the term.

Accordingly, when a keyword that can be called a keyword representing the tendency of the user 1 is calculated from the preference sentence according to the configuration of the tendency analysis module 900 , the image editing module 800 is determined according to the extracted keyword. By differentially applying an image editing technique to a rendered image based on a user's tendency, an image assigned as a visual effect can be generated. Therefore, it is possible to generate an image to which the editing technique (dimming, dissolve, zoom-in, zoom-out, fade) is differentially applied by reflecting the preference of the user 1 .

As described so far, the configuration and operation of the system for providing a high-speed rendering service for an image according to the present invention are expressed in the above description and drawings, but this is merely an example and the spirit of the present invention is not limited to the above description and drawings. It goes without saying that various changes and modifications are possible within the scope of the present invention without departing from the technical spirit of the present invention.

1: User 2: Central control server
100: condition setting module 200: file input module
300: rendering execution module 400: result providing module
500: result modification module 510: modification request unit
520: re-rendering unit 530: file providing unit
600: test rendering module 610: test file input unit
620: test rendering unit 630: test result providing unit
700: payment module 800: video editing module
900: propensity analysis module 910: term extraction unit
920: frequency matrix generator 930: transpose matrix generator
940: simultaneous matrix calculation unit 950: singular value calculation unit
960: final numerical calculation unit 970: key word extraction unit

Claims (8)

A system for providing a high-speed rendering service for an image, comprising:
a condition setting module configured to set rendering conditions including the number of frames from a user;
a file input module for uploading a rendering target image from a user;
a rendering execution module that renders the rendering target image according to the rendering condition to generate a rendered rendered image;
an image editing module that applies an image editing technique to the rendered image to give a visual effect;
A term extracting unit for extracting terms including an adjective and a verb after segmenting the preferred sentence into morpheme units as inputted from the user and analyzing and processing a preferred sentence for the video editing technique, and the term; A frequency matrix generation unit for generating a frequency matrix based on the frequency of occurrence in the preferred sentence, a transpose matrix generation unit for generating a transpose matrix by transposing the frequency matrix, and a term simultaneous based on the frequency matrix and the transpose matrix A simultaneous matrix calculation unit for generating a matrix and a sentence simultaneous matrix; a propensity analysis module including a final numerical value calculator for calculating a final numerical value based on a singular value, and a main word extracting unit for extracting a main word from the term according to the high and low of the final numerical value;
A result providing module that differentially applies the image editing technique to the rendered image to which the visual effect is given according to the user's tendency identified according to the extracted keyword and provides the result to the user;
The frequency matrix is
It is generated through the following Equation 1,
The transpose matrix is
It is generated through Equation 2 below,
The term simultaneous matrix is,
It is calculated through the following Equation 3,
The sentence simultaneous matrix is,
It is calculated through the following Equation 4,
The term singular value is
It is calculated through the following Equation 5,
The sentence singular value is
Rendering service providing system, characterized in that calculated through the following equation (6).
Equation 1,

Equation 2,

Equation 3,

Equation 4,

Equation 5,

Equation 6,

(here,

is the frequency matrix,

is the transpose matrix,

is the term concurrent matrix,

is the sentence simultaneous matrix,

Is

second term (

) entered into the system

second preferred sentence (

), the number of occurrences in

is the transpose matrix

line

column value,

is the term concurrent matrix

line

column value,

is the sentence concurrent matrix

line

column value,

is the number of generated terms,

is the number of preferred sentences input into the system,

is the term singular value,

is the sentence singular value,

is the maximum value among the eigenvalues of the term simultaneous matrix,

is the maximum value among the eigenvalues of the sentence simultaneous matrix) The method of claim 1,
The system is
a correction request unit receiving a re-rendering request for the rendered image from the user;
a re-rendering unit for re-rendering the rendered image according to the re-rendering request to generate a corrected image;
and a result correction module including a file providing unit that provides the corrected image to the user. The method of claim 1,
The system is
a test file input unit for receiving a test rendering image from the user;
a test rendering unit that renders the test rendering image to generate a test execution image;
and a test rendering module including a test result providing unit that provides the test execution image to the user.
The file input module,
A system for providing a rendering service, characterized in that the rendering target image is uploaded from the user provided with the test execution image. The method of claim 1,
The system is
Includes; a cost payment module receiving payment for the rendering from the user;
The rendering performing module,
As the payment of the cost is completed, the rendering service providing system is characterized in that the rendering target image is rendered. delete delete delete delete

Images