KR101623041B1 - System and method for managing markers coexisting mixed space, and the recording media storing the program performing the said method - Google Patents

Abstract

The present invention relates to a marker management system for managing markers located in both spaces in a mixed space in which a real space and a virtual space coexist in a virtual space and a real space according to a authoring technique for a context aware mobile AR, And a recording medium on which a program implementing the method is recorded. The present invention relates to a real space reference marker positioned at one point in a real space; A virtual space fiducial marker positioned in the virtual space and interlocked with the physical space fiducial marker and acting as an intermediary for enhancing the object at one point in the virtual space; A marker processing unit for generating an object mark marker for an object located in a real space or a virtual space on the basis of the actual space reference marker or the virtual space reference marker and processing the information on the object based on the generated object markers And a marker management system characterized in that the marker management system is provided. According to the present invention, two spaces can share information seamlessly in a mixed space where a real space and a virtual space coexist. Also, the rendering time of the image can be reduced in the augmented reality implementation and the like.

(AR), mixed space, VE (Virtual Environment) reference marker, RE (Real Environment) reference marker, CAMAR (Context Aware Mobile Augmented Reality) tag, marker interlock, scene graph,

Description

TECHNICAL FIELD The present invention relates to a marker management system and method for managing markers coexisting in a mixed space, and a recording medium on which a program embodying the method is recorded. }

The present invention relates to a marker management system and method for managing markers coexisting in a mixed space, and a recording medium on which a program implementing the method is recorded. More particularly, the present invention relates to a method and apparatus for creating a virtual space in a mixed space in which a virtual space and a virtual space coexist, which enables content authoring in a virtual space and a real space according to authoring technology for Context Aware Mobile Augmented Reality (CAMAR) A marker managing system for managing markers located in a space, a method thereof, and a recording medium on which a program implementing the method is recorded. In the above, context aware mobile augmented reality (CAMAR) authoring technology refers to a technology that allows mobile users to author multimedia contents in both a virtual space and an augmented reality environment through indoor / outdoor objects.

The research on the existing intelligent application service is centered on engineering, so it is limited to general users who are interested in well - being life care. Therefore, in order to realize the future well-being lifecare service that can be utilized in everyday life, user-oriented research that integrates user needs analysis, behavior modeling, user interface design, interaction, and ubiquitous computing is needed.

In recent years, with the advent of web 2.0-oriented Internet services such as facebook, youtube, and wikipedia, users have been able to freely publish and share multimedia contents, which has led to increased participation in cyberspace. Is still limited. However, as evidenced by the rapid growth of the market in the second life, which is a case study of virtual reality, it is clear that the boundaries between virtual and reality spaces gradually disappear. Therefore, it is necessary to preoccupy and lead the mobile augmented reality authoring technology which can present a new business model through fusion of real space and virtual space.

Recently, urban computing and social computing concepts are rapidly spreading and interest in community services is increasing. However, existing ubiquitous computing environments utilize only a part of the context such as the user 's profile or location, which limits the needs of community members. Therefore, it is necessary to develop a new type of social u-content authoring technology that can reflect the social relations and cultural background of the members so as to achieve the purpose of the community while meeting the needs of the members.

On the other hand, in the conventional augmented reality technology, mutual information is not shared between the virtual space and the real space. Therefore, in order to enhance the object in the virtual space as in the real space, it is necessary to calculate the position information and the orientation information of the object, and to extract the characteristic point of the object. Therefore, not only does it take much time to render the image, but it also requires a lot of time and manpower investment in recognition and tracking for enhancement.

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above problems, and it is an object of the present invention to provide a context aware mobile augmented reality (CAMAR) which enables mobile users to author multimedia contents in both virtual space and augmented reality environment through indoor / Augmented Reality), a marker management system that manages markers that are located in both spaces in a mixed space in which a real space and a virtual space coexist, The present invention also provides a recording medium on which a program for realizing a program for implementing the invention is recorded.

SUMMARY OF THE INVENTION The present invention has been made in order to accomplish the above object, and it is an object of the present invention to provide a real space reference marker positioned at one point of a real space. A virtual space fiducial marker located in a virtual space and interworking with the physical space fiducial marker and acting as a medium for enhancing an object at one point of the virtual space; An object mark marker for an object located in the virtual space or the real space based on the physical space fiducial markers or the virtual space fiducial markers, and processes the information on the object based on the generated object markers And a marker processing unit.

Preferably, the marker processing unit recognizes and tracks at least one object positioned in the real space or the virtual space using the object markers based on the actual space reference marker or the virtual space reference marker.

Preferably, the marker processing unit shares information about the object located on at least one space with reference to the real space reference marker or the virtual space reference marker.

Preferably, the information on the object to be processed by the marker processing unit includes information on the marker ID, information on the generation time of the marker, information on the position and orientation of the marker, information on the user, Information on the size of the object, information on the minutiae of the object, information on the texture of the object, and information on light illuminating the area where the object is located.

Preferably, the marker processing unit includes a mobile terminal for viewing the real space and generating the object markers; And a server for storing information about the object and information about the virtual space model, and providing the stored at least one information to the mobile terminal. More preferably, the mobile terminal includes a marker generation module for generating the object markers; A content authoring module for authoring content composed of at least one of audio, video, and text; A registration module for registering information on the ID of the object markers or information on the position and orientation of the object markers; A recognition module for recognizing the object; A tracking module for tracking the object; A sensing module for sensing a position and a direction of a person holding the mobile terminal; And an update module for acquiring and updating the list of object markers from the server based on the sensing result.

Preferably, the marker processing unit generates the virtual space reference marker or provides information about the object located in the virtual space to the real space reference marker. More preferably, the marker processing unit includes: a real space reference marker training unit for recognizing the actual space reference marker from the image of the pointed actual space and estimating the attitude; And a virtual space reference marker generation unit for generating the virtual space reference marker in the virtual space and linking the generated virtual space reference marker with the recognition space and the posture estimated reality space reference marker. Alternatively, the marker processing unit may include an object recognizing unit that recognizes an object included in the image of the pointing physical space; A virtual space reference marker extracting unit for extracting the virtual space reference marker having a maximum confidence value for the recognized object; A real space reference marker recognition unit for recognizing a reality space reference marker included in the image and estimating a posture; And an overlay implementation for overlaying information about the extracted virtual space reference marker on the recognized real space reference marker.

Preferably, the at least one of the physical space fiducial markers, the virtual space fiducial markers, and the object representation markers is linked with a marker provided in the same space or a marker provided in another space. And a sensing module for sensing movement of the marker provided in the space.

Preferably, the virtual space reference marker generator includes: a virtual space generator for generating a virtual space model having the same size as the physical space; And a corresponding object input unit for inputting a corresponding object corresponding to the object included in the image of the real space into the generated virtual space, and generates the virtual space reference marker to be connected to the input corresponding object. Also, the real space reference marker training unit may include a photographing unit for pointing the real space; An image acquiring unit acquiring an image of the pointing physical space; And a context information acquisition unit for acquiring context information including location information and orientation information of an object included in the acquired image, and recognizing the real space standard marker in consideration of the obtained image and the obtained context information, And estimate the posture.

Preferably, only the real space reference marker training portion or both the real space reference marker training portion and the virtual space reference marker generating portion are provided in the mobile terminal.

Preferably, the real space reference marker training unit further includes a user information collecting unit for collecting user information including at least one of a user's belonging, a keyword of interest, a preference, and information about the selected area, The actual space standard marker is recognized and the posture is estimated.

Preferably, the overlay implementer performs the overlay after matching the coordinate information of the extracted virtual space reference marker with the coordinate information of the recognized real space reference marker.

The virtual space fiducial marker extraction unit may extract a virtual space fiducial marker having the largest number of objects in the representative image, based on the coordinate information of the recognized objects, .

Preferably, the marker processing unit recognizes or tracks an object located at one point in the real space or the virtual space using information about the object. More preferably, the marker processing unit considers the amount of change in illuminance when recognizing or tracking the object.

According to another aspect of the present invention, there is provided an information processing method including the steps of: (a) acquiring position information and orientation information of an object located in the real space with reference to a real space reference marker positioned at one point of the real space; (b) selecting a virtual space model for reflecting the object; (c) generating an object located in the virtual space model using the acquired position information or the obtained orientation information on the basis of a virtual space reference marker on the selected virtual space model And a method for creating an object located in a space in a virtual space.

In the step (c), an object mark marker for representing the object in the virtual space model is generated, and the object is augmented in place of the generated object mark marker, thereby generating the object. More preferably, the step (c) may generate the object markers including the ID information of the object markers or the position information or the orientation information of the object markers, Content that is located is also augmented.

According to another aspect of the present invention, there is provided a method of detecting a pointing device, the method comprising: (a) recognizing a physical space reference marker representing an object positioned in the real space from an image of the pointing physical space; And (b) generating a virtual space reference marker in association with the real space reference marker and representing the object or the image including the object in a virtual space, And a step of interlocking with the spatial reference marker.

According to another aspect of the present invention, there is provided an image processing method comprising the steps of: (a) recognizing an object included in an image of a pointed physical space; (b) extracting a virtual space reference marker having a maximum confidence value for the recognized object; (c) recognizing the actual space reference marker included in the image and estimating the posture; And (d) superimposing information on the extracted virtual space reference marker on the recognized real space reference marker.

According to the present invention, the following effects can be obtained by managing markers located in both spaces in a mixed space in which a real space and a virtual space coexist for implementing a context aware mobile augmented reality (CAMAR) authoring technology. First, by managing two markers that work together in both spaces, the two spaces can share information seamlessly. Secondly, the virtual space fiducial markers can reduce the rendering time by sharing information about the objects provided in the real space, and the recognition process and tracking process for the enhancement can be omitted or proceed very quickly.

In addition, according to the present invention, a context aware mobile augmented reality (CAMAR) authoring technology capable of authoring multimedia contents in both a virtual space and an augmented reality environment through indoor / .

- Intuitive creation of content in virtual space utilizing the advantages of real space

- Effective / efficient production of contents of real space utilizing virtual space advantages

- Multiple users share content and information and collaborate to create

- A work that many people can participate at the same time

- Effective / efficient management and monitoring of contents and services in real space using virtual space

o CAMAR authoring core technology is expected to be able to secure internationally competitive technologies and intellectual property rights as a new technology that allows users to intuitively author desired contents by integrating virtual space and real space through mobile augmented reality. .

o CAMAR authoring core technology uses social relations and cultural context of members to reinforce social u-contents and enable mashups, thereby making it possible to transform existing technology-based mobile devices into human-centered helpers. At the same time, It is expected to play a key role in implementing the service.

o CAMAR authoring core technology, which allows users to participate and collaborate continuously, can mash up users by using their own contents as well as their own contents anytime and anywhere. Therefore, it is a tool for reusing and reconstructing various services and contents It is expected to be possible to use as.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same reference numerals are used to designate the same or similar components throughout the drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, the preferred embodiments of the present invention will be described below, but it is needless to say that the technical idea of the present invention is not limited thereto and can be variously modified by those skilled in the art.

In the embodiment of the present invention, the marker management system refers to a system for managing a target object through a reference marker located in each space in a mixed space in which a real world and a virtual world coexist .

When registering the target object in the virtual space, the marker management system can register the target object in the virtual space through a reference point of each space in which the information is shared. In this embodiment, the reference markers located in the virtual space are referred to as virtual space reference markers (i.e., VE (Virtual Environment) reference markers), the reference markers located in the real space are referred to as real space reference markers ), Respectively.

Further, the marker management system can manage the target object registered in the virtual space by using the object markers. The VE standard marker, the RE standard marker, and the object representation marker are all implemented as a CAMAR (Context Aware Mobile Augmented Reality) tag. The CAMAR tag is a concept that includes arbitrary 2D or 3D objects, as well as markers of fixed type, such as markers of ARToolKit. Therefore, in this embodiment, it is noted that the markers such as the VE reference marker, the RE reference marker, and the object markers include any 2D or 3D object in addition to a general marker (i.e., a marker in a fixed form).

More details of the marker management system will be described later. On the other hand, the marker management system can enhance a specific object in the virtual space in the real space through the interlocking of the VE reference marker and the RE reference marker.

The above-described marker management system will be described in detail with reference to FIG. 8 to FIG. 10 as "Context Aware Context (CAMAR) ", which allows mobile users to author multimedia contents in both a virtual space and an augmented reality environment through indoor / It is needless to say that two reference markers that enable contents authoring in both the virtual space and the real space can be managed together with the object according to the " Aware Mobile Augmented Reality ".

1 is a conceptual diagram of a marker management system according to a preferred embodiment of the present invention. Referring to FIG. 1, a marker management system 100 according to a preferred embodiment of the present invention includes a VE reference marker 110, an RE reference marker 120, an object marker 130, and a marker processing unit 140.

The VE fiducial marker 110 is a fiducial marker positioned in the virtual space and serves as an intermediary for enhancing an object at a certain position. In addition, the VE fiducial marker 110 plays a role in tracking recognition of at least one object located in the virtual space.

This function of the VE fiducial marker 110 allows the objects provided in the virtual space to be connected to each other. If objects are correlated, it is possible to render all the objects included in the virtual space by only recognizing and tracing one object through the VE reference marker 110. In this case, since it is not necessary to calculate the positions and orientations of the remaining objects except for one, it is possible to reduce the rendering time and to implement a scene graph that can systematically manage complicated virtual environments.

The VE fiducial markers 110 are connected to the RE fiducial markers 120. Accordingly, the VE reference marker 110 can be directly mapped based on the real environment information. At this time, it is preferable that the VE reference marker 110 is capable of recognizing and tracking not only vision but also hybrid.

As described above, the VE reference marker 110 may be implemented as a CAMAR (Context Aware Mobile Augmented Reality) tag in this embodiment. In this embodiment, the registered reference point is a set of contexts used for registering the target object in the virtual space in the real space, and defines this as a CAMAR Tag.

CAMAR Tag, which stores real objects in virtual space, is not only a reference point for viewing the augmented contents on other objects around the target object without a separate recognition process or tracking process, By registering through the CAMAR Tag, other users can share the virtual space, and the information can also act as a sharing point that is shared, and as a context-based search key in the context contained in the CAMAR Tag.

The CAMAR tag includes information such as a model context containing information about an object provided in the virtual space, a place where the object is created (e.g., position, orientation, etc.) and timing A tag data model including an environment context, a user context containing information on a user, and a contents context containing information on contents used in a work, .

The model context includes a global coordinate system including an ID of a tag representing an object, a local coordinate system including information on coordinates of the object, a scale including information on the size of the object, Model information including information such as a pose matrix, information on 3D points (3D points), information on 3D lines (3D lines) and other information, feature point information and edge information edges, texture information, and information for an ID including other information.

However, even for the same object, since the minutiae points for the ID can be obtained differently according to the time or illumination information when the CAMAR tag is generated, the time or illumination information at the time of recognition / tracking is used as an index, . That is, the ID information is information that considers the environmental context (e.g., optical information and time information) at the time of tag generation.

The environmental context includes information that can be obtained from the environment when the CAMAR Tag is created, such as location, direction, light, time, and so on.

The user context includes information based on content authoring such as information on the owner, a community belonging to the user, a keyword indicating intention, and the like.

The content context includes various contents such as text, audio, video, 2D content, and 3D content.

In the present embodiment, the CAMAR tag described above is defined as a tag that can be recognized and tracked by a hybrid method between a sensor and a vision by attaching various types of sensors such as an AR marker and a visual sensor, IR, and RFID. For example, the CAMAR tag may be a combination of a visual sensor, an active sensor, a passive sensor, and the like. The AR marker is a concept including a marker representing a virtual object in an augmented reality environment, a marker frame surrounding the marker and displaying positional information and orientation information of the marker.

If the VE standard marker 110 is implemented as a CAMAR tag having such interoperability, it is possible to more easily implement interworking with the RE standard marker 120.

The RE reference marker 120 is a marker positioned in the real space and performs the same function as the VE reference marker 110. [ The RE reference marker 120 may be implemented as a CAMAR tag in the same manner as the VE reference marker 110.

The VE reference marker 110 and the RE reference marker 120 described above are smart objects respectively present in the virtual space and the real space, and function to continuously connect the virtual space and the real space. The VE reference marker 110 and the RE reference marker 120 share orientation information with the position information of the same object located in each space. Accordingly, the VE reference marker 110 and the RE reference marker 120 can continuously represent corresponding objects in the virtual space and the real space.

Meanwhile, when the object existing in the virtual space but not in the real space is to be enhanced in the real space, the position information and the orientation information of the object when the VE reference marker 110 is used as the reference are added to the VE reference marker 110 You can do that. As described above, in the present invention, it is possible to freely enhance the object in the real space only by interlocking the VE reference marker 110 and the RE reference marker 120 existing in the virtual space and the real space without extracting the marker or feature point have.

The object markers 130 are markers for representing objects provided in a real space or a virtual space. The object markers 130 may be implemented as CAMAR tags in the same manner as the VE reference markers 110 and the RE reference markers 120 and may enhance the objects in virtual spaces or predetermined positions in the real space.

The object markers 130 provided in the virtual space or the real space are organically connected to each other through the CAMAR tag. That is, by sharing position information and azimuth information, neighboring objects positioned within a predetermined distance based on a specific object are interlocked. Thus, even if the environment is changed, the object markers 130 facilitate recognition and tracking of other objects from one object through mutual organic relationships.

On the other hand, the object markers 130 may be implemented by a marker having a sensing module and a networking module. The object markers 130 having such a configuration also allow the object to be displayed to be linked with other objects and can implement a spatial representation.

The marker processing unit 140 processes the objects located in the virtual space or the real space based on the VE reference marker 110 or the RE reference marker 120.

The marker processing unit 140 may include a mobile device and a virtual world server in this embodiment. The mobile terminal is used to create the object markers 130 or the object markers 130 to view the augmented reality space. In addition, the virtual model server 350 is used to store information about the object markers 130 and information about the virtual space model. However, the marker processing unit 140 is not limited to this configuration, and may be implemented by a single terminal or a single server. Hereinafter, the configuration and operation of the marker processing unit 140 will be described with reference to FIG.

The mobile terminal 300 may be implemented as a laptop, an UMPC, a PDA, a smart phone, or the like.

The virtual model server 350 is a server that processes the object markers 130 in cooperation with the mobile terminal 300. The virtual model server 350 adds the CAMAR tag as a node to the correct position of the virtual space model based on the coordinate information of the model context (i.e., the global coordinate and the local coordinate) when a CAMAR tag registration request is made.

In addition, when there is an update request for the CAMAR tag, the virtual model server 350 retrieves the CAMAR tag list by performing contextual indexing based on the key information of the user context using the environment context. Model information (particularly, minutiae points) of the object is created depending on the environmental context value such as roughness when the model information of the object is created. Accordingly, when the object is recognized / traced, the object model information corresponding to the illuminance value is transferred and processed. In this embodiment, it is based on performing contextual indexing using environment context based on context key information.

The mobile terminal 300 and the virtual model server 350 operate as follows in order to create, register, manage, and utilize objects using the CAMAR tag. We explain each in turn.

① Object creation and registration

First, the sensing module 305 acquires an environmental context (e.g., location, direction, etc.) of the user including the object. Thereafter, the CAMAR tag update module 306 receives the model view from the virtual space model provision module 353. [ At this time, it is preferable that the model view is a model view of a virtual space coinciding with a view of a real space being visually confirmed through the mobile terminal 300.

The global coordination matching module 303 aligns the provided model view with the view of the real space (coordinates the coordinate system). Then, the local coordinate matching module 304 obtains the data that comes in relation to the global coordinate of the target object of interest, and sets the local coordinate. The local coordinate matching module 304 obtains model information (e.g., recognition points, tracking points) for an object through interaction of a stylus pen input or the like with an object on the screen.

Then, the CAMAR tag generation module 301 generates the CAMAR tag including the acquired environment context and the model context. Preferably, the CAMAR tag generation module 301 generates an object as a CAMAR tag, including a user context and a content context through authoring. When the CAMAR tag includes the content context, the process of generating the content context by the content authoring module 302 precedes the process of generating the CAMAR tag.

Thereafter, the CAMAR tag registration module 352 registers the generated CAMAR tag. In accordance with this driving of the CAMAR tag registration module 352, the generated CAMAR tag is stored in the virtual model server 350. [

② Object Management

The virtual model server 350 stores the CAMAR tag generated from the mobile terminal 300 in the database. Preferably, the virtual model server 350 stores the CAMAR tag in the database along with the applicable virtual model. Here, the applicable virtual model is a model view that is provided from the virtual space model provision module 353 by the CAMAR tag update module 306 described above. However, the virtual model applicable in the present embodiment is not limited to this. That is, the virtual model that can be applied in this embodiment can be defined as any model view that can reflect the CAMAR tag including the model view.

When the CAMAR tag registration request message is received from the mobile terminal 300, the CAMAR tag registration module 352 refers to the model context, in particular, the global coordination and the local coordinate, and stores the CAMAR tag in the applicable virtual model as a node. In this case, the node can be defined to be stored in association with the reference marker of the virtual model using the model context.

The mobile terminal 300 transmits a CAMAR tag update message to the virtual model server 350 after the CAMAR tag registration request. Then, the CAMAR tag list retry module 351 returns the CAMAR tag list by context indexing using the environment context of the CAMAR tag as a key.

③ Utilization of objects

First, the sensing module 305 of the mobile terminal 300 acquires the environment context and the user context. Then, the CAMAR tag update module 306 acquires the virtual model through the CAMAR tag list retry module 351. [ At this time, the ROI selection module 314 may extract the region of interest (ROI) from the entire region of the virtual model, and then display only this portion on the mobile terminal 300.

Thereafter, the CAMAR tag recognition module 312 recognizes the target object using the acquired minutiae of the model context, and the CAMAR tag tracking module 313 tracks the target object. Thereafter, the content enhancement module 311 augments the content.

Referring back to FIG.

The marker processing unit 140 may include at least one object markers (not shown) associated with the RE fiducial markers 120 or the RE fiducial markers 120 such that the CAMAR tag is present in both spaces (real space and virtual space) 130 to generate a VE criterion marker 110 that provides information about the marker group. More specifically, the marker processing unit 140 generates a model corresponding to the image in the virtual space when at least one object is displayed in the real space through the RE marker 120 and the object markers 130, An object is input to the model generated through the marking marker 130 and the input object is connected to the VE reference marker 110 generated corresponding to the RE reference marker 120 so that the VE reference marker 110 is displayed It is possible to provide information on the same image as the real space.

The marker processing unit 140 performs a function of authoring contents in one space by using the VE reference marker 110 or the RE reference marker 120. [

Meanwhile, the marker processing unit 140 may further include a power unit and a controller for smooth driving. The power supply unit supplies power to all components constituting the marker processing unit 140, and the control unit controls the overall operation of all the components constituting the marker processing unit 140.

Next, the marker processing unit 140 generates an object in the virtual space using the VE reference marker 110 and the RE reference marker 120 in the virtual space. 4 is a flowchart illustrating an object generating method according to a preferred embodiment of the present invention. The following description refers to Fig.

First, the sensing module 305 acquires position information and orientation information of an object in a real space (S400). Thereafter, the CAMAR tag update module 306 selects a virtual space model for reflecting the object in the real space (S410). Then, the CAMAR tag generation module 301 generates an object representation marker 130 for the object so that the object can be reflected in the selected virtual space model (S420). It is possible to provide the virtual space with objects in the real space through the generated object markers 130. [

Thereafter, the global coordinate matching module 303 registers the ID of the object markers 130 (S430), and the local coordinate aligning module 304 registers the coordinate information of the object markers 130 (S440).

Thereafter, if there is content to be augmented together with the object, the content authoring module 302 writes the content together with the object in the virtual space model (S450). When authoring of the object and the content is completed, the virtual space model is completed with the object reflected (S460).

5 is a reference diagram for explaining an object generating method according to a preferred embodiment of the present invention. FIG. 5 is matched to each step of FIG. 4 as follows.

5 (a) is a view of the actual space image. In FIG. 5 (a), information about the target object 500 is obtained through step S400. 5 (b) is an image for the virtual space model. The target object 500 is reflected in the place of the object markers 510 in FIG. 5 (b), and FIG. 5 (b) is determined through step S410.

FIG. 5C shows a process of obtaining the orientation information of the target object 500. FIG. Information on the target object 500 obtained from FIG. 5C is registered as information on the object markers 130 through step S440. 5D is an image of the virtual space model in which the object 420 and the contents 430 are reflected. Through steps S400 to S460, a virtual space model image as shown in (d) of FIG. 5 can be obtained.

Next, a method of generating the VE reference marker in which the marker processing unit 140 interlocks with the RE reference marker (the marker representing the object provided in the physical space) will be described. FIG. 6 is a flowchart illustrating a method of generating a VE reference marker according to a preferred embodiment of the present invention. The following description refers to Fig.

First, the photographing unit 321 points the real space RE using the camera (S600). Thereafter, the image obtaining unit 322 obtains a plurality of image scenes for the pointed actual space (S610a). In operation S620a, the context information obtaining unit 323 obtains context information including location information and azimuth information of at least one object included in the acquired image scenes.

Meanwhile, the user information collection unit 324 collects profile information of the user (S610b). The user's profile information collection can be done through user's information input. At this time, the profile information of the user collected by the user information collecting unit 324 includes information on belonging, interest keyword, preference, area selection, and the like.

Thereafter, the RE reference marker training unit 325 recognizes the ID of the RE reference marker from the image scenes obtained in consideration of the collected profile information of the user, the acquired context information, and the like, and estimates the posture thereof (S630).

In the embodiment of the present invention, objects in the virtual space are provided in the same manner as objects in the real space. In addition, the objects provided in each space are arranged according to a predetermined positional relationship. Therefore, it is necessary to load a virtual space model having the same size as the real space before generating the VE reference marker. It is also necessary to input an object in a virtual space corresponding to each object in the real space.

After step S630, the virtual space creating unit 326 creates a virtual space model having the same size as the real space (S640). Thereafter, the corresponding object input unit 327 inputs corresponding objects corresponding to the objects in the real space to the virtual space model using the context information (S650). In the above, the context information is a concept including position information and orientation information of an object provided in the real space. Therefore, if the coordinate system information (position information and orientation information) of the object in the real space and the object in the virtual space are matched, the corresponding object can be input.

When the virtual space having the same state as the real space is completed, the VE marker generating unit 328 generates a VE reference marker corresponding to the RE reference marker (S660). At this time, the VE marker generating unit 328 generates the VE reference marker to be associated with all the input corresponding objects. As a method of connecting the corresponding object to the VE reference marker, a method in which the web information is linked to hypertext can be used.

The VE reference marker generated in step S660 includes recognition information of an RE reference marker, posture estimation information of an RE reference marker, connection information on a corresponding object, and the like. Here, the recognition information of the RE reference marker includes the ID of the RE reference marker. The posture estimation information of the RE reference marker includes coordinate system information including the position information of the RE reference marker and the orientation information. The connection information for the corresponding object includes position information and orientation information of at least one corresponding object provided in the virtual space model. On the other hand, the VE reference marker may further include a scale measurement index.

Meanwhile, the photographing unit 321 to the VE marker generating unit 328 for performing the respective steps shown in FIG. 6 are provided in the marker processing unit 140. In an embodiment of the present invention, the marker processing unit 140 may be a computer or a mobile terminal. A computer can be a server, for example, and a mobile terminal, for example, a mobile phone, a PDA, or the like. However, it is also possible that the marker processing unit 140 is not a single device but a multiple device (e.g., a server computer mounting a mobile terminal) or includes a plurality of devices as mentioned in FIG. For example, the marker processing unit 140 includes the mobile terminal 300 including the photographing unit 321 to the RE reference marker training unit 325 and the virtual space generating unit 326 to the VE reference marker generating unit 328 Server computer 350 as shown in FIG.

Next, a method of using the VE reference marker 110 by the marker processing unit 140 will be described. FIG. 7 is a flowchart illustrating a method of utilizing a VE reference marker according to a preferred embodiment of the present invention. The following description refers to Fig.

First, the photographing unit 321 views the real space RE (S700). Then, the image acquisition unit 322 acquires the video scenes for the real space (S710), and the context information acquisition unit 323 acquires at least one RE object (object included in the real space) included in the video scenes Context information is obtained (S720). At this time, it is also possible to collect profile information of the user through the user information collection unit 324.

Thereafter, the object recognition unit 331 recognizes at least one object included in the video scenes based on the acquired context information (S730). If the VE reference marker list providing unit 332 provides the available VE reference marker list stored in the memory unit in operation S740, the VE reference marker extracting unit 333 extracts the VE reference marker having the maximum confidence value (S750). In the embodiment of the present invention, the VE reference marker having the maximum reliability value is a VE reference marker that represents a RE video scene having the largest number of objects having the same context information based on the context information (position information and orientation information) Lt; / RTI >

Thereafter, the RE reference marker recognition unit 334 recognizes the RE reference marker included in the video scenes and estimates the posture thereof (S760). Preferably, the RE reference marker recognition unit 334 also measures the scale. Then, the overlay implementing unit 335 acquires information on the extracted VE reference marker and overlayes the contents of the VE reference marker in place of the RE reference marker 120 (S770). In the embodiment of the present invention, coordinate information (position information and azimuth information) of the VE fiducial markers and coordinate information of the RE fiducial markers are matched with each other before the implementation of such an overlay.

The following applications are also possible if the marker management system described with reference to Figures 1 to 7 is used.

1. CAMAR block

Blocki (blogi) is a concept that combines an existing blog (blog) with a wiki. Blogs are recorded as log entries on a daily basis and are authorized to write to only one or a few people. These blogs include text, photos, music, flash, videos, and can be commented or trackbacked.

On the other hand, wiki is a collaborative software, and everyone has the advantage of writing. In some, you write two or more words and write the first letter of each word in capital letters. This wiki comes from the fact that several different intellectual beings join forces to organize the expansion, identification and connection of content through interaction. It is desirable that the wiki be relatively strictly examined and reviewed through mutual criticism, and it is also possible to co-create collective intelligent results at the same time as satisfying oneself.

If the marker management system according to the present invention is applied, an augmented reality collaboration software that enables CAMAR bloc i, that is, a common goal in a community, to create a new experience by recording and managing experience for the community is also possible.

The above-described embodiments of the present invention can be embodied in a general-purpose digital computer that can be embodied as a program that can be executed by a computer and operates the program using a computer-readable recording medium. The computer readable recording medium may be a magnetic storage medium (e.g., a ROM, a floppy disk, a hard disk, a magnetic tape, or the like), an optical reading medium (e.g., a CD-ROM, a DVD, And a carrier wave (e.g., transmission over the Internet).

Next, referring to FIG. 8 to FIG. 10, a context aware mobile augmented reality (CAMAR) system capable of authoring multimedia contents in both a virtual space and an augmented reality environment through indoor / Authoring core technology ".

"Context Aware Mobile Augmented Reality (CAMAR) authoring technology for indoor and outdoor enhancement and mashup of social u-contents coexisting in intelligent mixed space" is a social interaction between community members in USS (Ubiquitous Smart Space) And cultural backgrounds, we define social U-content as a CAMAR core technology that enhances and mashups the intralet object and enables selective sharing and continuous collaboration among its members.

The social u-contents generated by using the integrated context including the social relations and cultural background among the community members in the intelligent mixed space (USS) in which the virtual space and the real space coexist include characters, logos, signs and buildings We will develop a context aware mobile augmented reality authoring core technology that seamlessly integrates indoor and outdoor objects, reconstructs social u-contents in the form of participating and constant collaboration, and enables selective sharing.

8 is a diagram showing a conceptual diagram of the CAMAR authoring core technology.

The proposed authoring technique is a mobile augmented reality - based authoring technology that can create contents in both virtual space and real space. In the virtual space, it is possible to manipulate / modify / combine / move / copy desired multimedia contents based on the GUI. In the real space, real objects are recognized and tracked through mobile augmented reality to enable authoring functions such as acquisition / selection / movement / transformation / combination / copy of multimedia contents. In addition, in the real space, the contents in the virtual space can be acquired and authored through the augmented reality, and the contents changed in the real space are reflected again in the virtual space. In addition, it is possible to author new contents by searching and utilizing contents authored by other users.

9 is an example of the concept of the CAMAR authoring technique in an indoor environment. In this picture, the work in the real space, the smart home, and the work in the virtual smart home are performed at the same time. Two users in the virtual space are authoring the robot taekwon v. At the same time, the users in the virtual space are writing the same robot taekwondo. To do this, we recognize Smart Table in real space and acquire contents existing in virtual space and author intuitively. In a virtual space, authoring works with other users even if they are not actually located based on the contents that are changing in the real space.

The proposed authoring method consists of augmented reality authoring, virtual reality authoring, and content synchronization module. The real space work is located on the mobile device and performs the authoring function necessary for each user. In addition, by utilizing the place where the user is located, the related content is linked with the virtual space. The virtual space authoring department can access the users through the internet and authoring is performed through the GUI. And, the content synchronization module performs sharing and coherence maintenance of the generated authoring events in the two storage spaces.

To this end, the augmented reality work consists of an object recognition and tracking module, an object selection module, an object storage module, a rendering module, and a virtual object management module. The object recognition and tracking module recognizes and tracks the object by analyzing the image input through the camera image. The object storage module stores various multimedia contents selected according to the location of the user and the user. Virtual reality authoring consists of a GUI-based interface module, an object manager, an object manipulation module, a content controller, and a renderer. The content synchronization module includes a server synchronization module and an augmented reality synchronization module. Fig. 10 shows a structural diagram of the proposed fabrication technique.

≪ Differentiation of the present invention >

Existing work is completely separated between virtual space work and real space work. This study is a technology that can mix two spaces and author in both directions.

- Mapping technique between real space and virtual space through CAMAR tag (smart objects and objects located inside / outside of signboard, building shape, etc.).

Content Synchronization Technique for Event - Context - based Real - space and Virtual Space.

- Authoring technology for creating virtual space contents (3D models, services, sensors) using augmented reality and real objects.

- Object recognition / tracking technology of real space using virtual space and model.

- Content authoring technology in augmented reality environment using virtual space and model.

- Authoring technology that can create contents simultaneously in both augmented reality space and virtual space.

- Authoring technology that searches and utilizes contents authored by other users to create new contents.

- Authoring technology that can create content in real environment and object in remote environment in mobile environment.

- A group recommendation technique reflecting the social relationships and cultural backgrounds of multiple users.

- Content Response Generation Method Reflecting Social Relationship and Cultural Background of Multiple Users.

Specific technology development according to the present invention is as follows.

1. Develop social u-content enhancement and interaction technologies that harmonize and coordinate opinions among community members

A. USS service scenario and user interface design based on user community

- CAMAR service design and usability evaluation

B. Sensor-fusion-based outdoor object recognition and tracking technology for social u-content visualization

- Marker tracking and matching technology using character recognition

- At a resolution of 320 × 480 image, character recognition rate is 80% or more, marker recognition rate is 90% or more, tracking rate is 20fps

- Model-based outdoor building tracking and matching technology through information fusion of camera external sensor (GPS) (tracking speed 10 fps or more, tracking error within 15 pixels)

C. Collaboration and CAMAR mashup based on location and preference among users in virtual and real space communities

- CAMAR community creation and group recommendation technique based on place and preference

- community creation within 3 seconds, recommendation accuracy over 65%

- Social u-content information level adjustment and filtering technology based on personal situation and selection (more than 70% feasibility, less than 0.2 seconds of filtering time)

- Virtual environment configuration and access technology in real environment

D. Context-aware social u-content intelligence technology to collect and understand contexts to support the interaction of community members

- Acquisition of user feedback based on biological signal context of CAMAR user and social u-content context recognition technology

- Personalized decision technology of social u-content according to CAMAR user's context

- CAMAR user's context-based social u-content

E. CAMAR authoring platform and application system

- Formalization and design of CAMAR authoring platform development process that can operate on mobile phone

- Social u-content information registration, insertion, and positioning technology for mobile phone based u-content authoring

- Physics engine suitable for mobile phone

- A technique of providing physical attributes of objects for realistic reaction of objects in augmented reality environment and collision detection technology of augmented virtual objects

2. CAMAR service mash-up technology for service acquisition, enhancement and reconstruction in the community

A. Service scenario and user interface design for CAMAR service mashup

- CAMAR mashup service design and usability evaluation

B. Support for social u-content visualization and interaction Outdoor smart object recognition and tracking technology

- Logo and signboard tracking and matching technology using character and 2D barcode recognition (85% character recognition rate, 90% bar code recognition rate, 25fps tracking speed, 15fps tracking error, 10 pixel tracking error)

- Camera external sensor and model fusion based outdoor building tracking and matching technology (tracking speed 15fps or more, tracking error within 10 pixels)

C. Social Context-based Collaboration and CAMAR Mashup Technology in Community

- Information sharing and intervention techniques within the CAMAR community based on social context (less than 2 seconds for community creation time and more than 75% for group recommendation accuracy)

- Service filtering configuration technology according to user choice and social relations in CAMAR community (within 0.15 second of filtering execution time, accuracy and feasibility of provided information more than 75%)

- Information sharing level synchronization technology in mixed environment

D. CAMAR Interaction technique among multiple social u-contents based on social context of community members

- Acquisition of multi-modal feedback by CAMAR users and social context recognition technology of social u-contents

- CAMAR Situation determination technology that supports interaction of social u-content according to user's context

- Social context-based autonomous action planning technology of multi-social u-content

E. CAMAR authoring platform and application system

- Various features and functionality enhancements required by the CAMAR authoring platform (90%)

- Mobile phone based u-content authoring technology

- u-content sharing technology through simultaneous access of physics engine

3. Developing service recommendations, experience recording and enhancement techniques for community members

A. CAMAR Service Scenario Design Based on Social Relationships and Cultural Context

- CAMAR service design and usability evaluation for service recommendation, experience recording and enhancement

B. Supporting social u-content visualization and interaction Many smart object recognition and tracking technology in indoor

- Object tracking and matching technique using handwritten characters and 2D barcode recognition (Character recognition rate over 90%, resolution of barcode recognition 95% or more, tracking rate 25 fps or more in resolution 320 × 480 image)

- Multi-object tracking and matching technique based on camera external sensor and model convergence (tracking speed is over 15 fps, tracking error is within 7 pixels)

C. Social u-content collaboration and CAMAR mashup technology based on cultural context

- CAMAR community dynamic configuration based on cultural context and community intervention technology (community generation time less than 1 second, group recommendation accuracy more than 80%)

- Event-based mashup service setup, profile learning and filtering technology in CAMAR community

- Virtual environment DB construction technology based on context history of real environment

D. Behavior Generation Techniques for Social u-content that Acts Autonomously Based on Experience Between Multiple Social u-Contents

- Acquisition of emotional context of community members using social u-content and cultural context recognition technology of u-contents

- Adaptive Situation Judgment Technique of Social u-contents based on sharing experiences among CAMAR users in the community

- Social context-based autonomous behavior generation technology of multiple social u-contents

E. CAMAR authoring platform and application system

- Provide a widget that allows flexible creation of social u-contents according to the context information of the community and improve flexibility (80%)

- Smart object and real object based authoring technology

- Expansion of physics engine with various objects such as rigid body and particle

4. Development of CAMAR authoring core technology for continuous collaboration based on social relations and cultural context of community members

A. CAMAR Service Scenario Design Based on Social Relationships and Cultural Context

- CAMAR authoring tool UI design and practical application design strategy for continuous collaboration among multiple users

B. Integrated recognition and tracking technology for social u-content visualization and authoring

- Object tracking and matching technique using handwritten characters and 2D barcode recognition (95% character recognition rate, 95% marker and logo recognition rate, and 25fps tracking speed at resolution 320 × 480 image)

- Multi-object tracking and matching technique based on camera external sensor and model fusion (tracking speed is 20 fps or more, tracking error is within 5 pixels)

C. Social u-content based on socio-cultural context in virtual and realistic public space, collaboration and CAMAR mashup technology

- CAMAR community extension and intervention technology based on socio-cultural context (community creation within 1 second, recommendation accuracy over 90%)

- Service composition and classification technology for social u-content authoring in CAMAR community (more than 90% feasibility, less than 0.05 seconds of filtering time)

- Mixed environment information synchronization and integration technology

D. Self-growing technology of social u-content that learns and reflects social and cultural context among community members

- social feedback acquisition and learning skills of community members using social u-content

- Learning technology for the adaptive situation judgment of the social u-contents according to the reaction of the social u-contents of the CAMAR users in the community and the feedback of the users

- Evolutionary action planning technology that enables learning of social agents by sharing social and cultural experiences

E. CAMAR authoring platform and application system

- Increase productivity of the authoring platform (80%), which minimizes social u-content authoring time and is easy to use by ordinary users.

- Collaborative authoring techniques for content of multiple CAMAR users

- commercialization through application scenarios, application of commercially available physics engine

It will be apparent to those skilled in the art that various modifications, substitutions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. will be. Therefore, the embodiments disclosed in the present invention and the accompanying drawings are intended to illustrate and not to limit the technical spirit of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments and the accompanying drawings . The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

The present invention provides a context aware mobile augmented reality (CAMAR) authoring technology that enables mobile users to author multimedia contents in both a virtual space and an augmented reality environment through indoor / outdoor objects, A marker management system and method for managing markers which are located in both spaces in a mixed space where a real space and a virtual space coexist for technology implementation, and a recording medium on which a program implementing the method is recorded do.

According to the present invention, the two spaces can share information seamlessly by managing two markers that work together in the two spaces. In addition, the virtual space fiducial markers can reduce the rendering time by sharing information about objects provided in the real space, and the recognition process and tracking process for the enhancement can be omitted or can be performed very quickly.

The present invention relates to a context aware mobile augmented reality technology and is suitable for ubiquitous computing using a mobile terminal. Considering that augmented reality technology has been widely applied in medicine, industry, entertainment, military field, and the main user interface is changing from HMD (Head Mounted Display) to mobile, It will be worth using.

1 is a conceptual diagram of a marker management system according to a preferred embodiment of the present invention;

FIG. 2 is a table defining a CAMAR tag according to a preferred embodiment of the present invention.

3 is a diagram illustrating an internal configuration of a marker processing unit according to a preferred embodiment of the present invention.

4 is a flowchart illustrating an object generating method according to a preferred embodiment of the present invention.

5 is a reference diagram for explaining an object generating method according to a preferred embodiment of the present invention,

6 is a flowchart illustrating a method of generating a VE fiducial marker according to a preferred embodiment of the present invention.

FIG. 7 is a flowchart illustrating a method of utilizing a VE reference marker according to a preferred embodiment of the present invention.

8 is a view showing a conceptual diagram of a CAMAR authoring core technology,

Figure 9 is an illustration of the concept of CAMAR authoring technology in an indoor environment,

Fig. 10 is a diagram showing a structure of a proposed fabrication technique.

Description of the Related Art

100: marker management system 110: VE reference marker

120: RE reference marker 130: object marker

140: marker processing section

Claims (27)

A physical space fiducial marker located at a point in the real space; A virtual space fiducial marker located in a virtual space and interworking with the physical space fiducial marker and acting as a medium for enhancing an object at one point of the virtual space; An object mark marker for an object located in the virtual space or the real space based on the physical space fiducial markers or the virtual space fiducial markers, and processes the information on the object based on the generated object markers The marker processing section The marker management system comprising: The method according to claim 1, Wherein the marker processing unit recognizes and tracks at least one object located in the real space or the virtual space using the object markers based on the actual space reference marker or the virtual space reference marker. . The method according to claim 1, Wherein the marker processing unit shares information about the object located on at least one space with reference to the real space reference marker or the virtual space reference marker. The method according to claim 1, The information on the object to be processed by the marker processing unit includes information on the marker ID, information on the generation time of the marker, information on the position and orientation of the marker, information on the user, information on the content used for authoring, Wherein the marker management information includes at least one of information on a size of the object, information on a feature point of the object, information on a texture of the object, and information on light illuminating an area where the object is located. The method according to claim 1, The marker processing unit, A mobile terminal for viewing the real space and generating the object representation marker; And A server for storing information about the object and information about the virtual space model and providing the stored at least one information to the mobile terminal, The marker management system comprising: 6. The method of claim 5, The mobile terminal comprises: A marker generation module for generating the object representation marker; A content authoring module for authoring content composed of at least one of audio, video, and text; A registration module for registering information on the ID of the object markers or information on the position and orientation of the object markers; A recognition module for recognizing the object; A tracking module for tracking the object; A sensing module for sensing a position and a direction of a person holding the mobile terminal; And An update module for acquiring and updating the list of object markers from the server based on the sensing result, And a marker management system. The method according to claim 1, Wherein the marker processing unit generates the virtual space reference marker or provides the information on the object located in the virtual space to the real space reference marker. 8. The method of claim 7, The marker processing unit, A real space reference marker training unit for recognizing the real space reference marker from the image of the pointed actual space and estimating the attitude; And A virtual space reference marker generation unit that generates the virtual space reference marker in the virtual space, and associates the generated virtual space reference marker with the recognized and attitude estimated real space reference marker, The marker management system comprising: 8. The method of claim 7, The marker processing unit, An object recognition unit for recognizing an object included in the image of the pointing physical space; A virtual space reference marker extracting unit for extracting the virtual space reference marker having a maximum confidence value for the recognized object; A real space reference marker recognition unit for recognizing a reality space reference marker included in the image and estimating a posture; And An overlay implementation unit for overlaying information about the extracted virtual space reference marker on the recognized real space reference marker, The marker management system comprising: 8. The method of claim 7, Wherein the real space fiducial marker or the virtual space fiducial marker comprises: A networking module for interlocking with a marker provided in the same space or a marker provided in another space; And A sensing module for sensing movement of the marker provided in the space, And a marker management system. 9. The method of claim 8, Wherein the virtual space fiducial marker generation unit comprises: A virtual space generation unit for generating a virtual space model having the same size as the real space; And A corresponding object input unit for inputting a corresponding object corresponding to the object included in the image of the real space into the generated virtual space; / RTI > And generates the virtual space reference marker to be connected to the input corresponding object. 9. The method of claim 8, The real space reference marker training unit includes: A photographing unit for pointing the real space; An image acquiring unit acquiring an image of the pointing physical space; And Acquiring context information including position information of the object included in the acquired image and orientation information, / RTI > Wherein the real space standard marker is recognized in consideration of the obtained image and the obtained context information, and the posture is estimated. 9. The method of claim 8, Wherein the real space reference marker training unit, or both the real space reference marker training unit and the virtual space reference marker generation unit are provided in the mobile terminal. 13. The method of claim 12, The real space reference marker training unit may further include a user information collection unit for collecting user information including at least one of a user's belonging, a keyword of interest, a preference, and information about the selected area, Wherein the real space standard marker is recognized and the posture is estimated. 10. The method of claim 9, Wherein the overlay implementer performs the overlay after matching the coordinate information of the extracted virtual space reference marker with the coordinate information of the recognized real space reference marker. 10. The method of claim 9, The virtual space fiducial marker extraction unit extracts a virtual space fiducial marker having the most information on the representative image with the object having the same information based on the coordinate information of the recognized objects with the virtual space fiducial marker having the maximum confidance value . 5. The method of claim 4, Wherein the marker processing unit recognizes or tracks an object located at one point in the real space or the virtual space using information on the object. 18. The method of claim 17, Wherein the marker processing unit considers the amount of change in illuminance when recognizing or tracking the object. A method of operating a terminal for viewing an augmented reality, (a) acquiring positional information and orientation information of an object located in the real space with reference to a real space reference marker positioned at one point of the real space; (b) selecting a virtual space model for reflecting the object; (c) generating an object located in the virtual space model using the acquired position information or the obtained orientation information on the basis of a virtual space reference marker on the selected virtual space model Wherein the object is located in a virtual space. 20. The method of claim 19, Wherein the step (c) comprises: generating an object mark-up marker for representing the object in the virtual space model, and enhancing the object in a position of the generated marker mark in the virtual space model; A method of creating an object in a virtual space. 21. The method of claim 20, Wherein the step (c) includes: generating the object markers including the ID information of the object markers or the positional information or the orientation information of the object markers, and the content positioned at a predetermined position from the object The object being located in the virtual space. An operation method of a marker management system for managing a marker for expression of an augmented reality, (a) recognizing a physical space reference marker representing an object located in the physical space from an image of the pointing physical space and estimating a pose; And (b) generating a virtual space reference marker in association with the real space reference marker in the virtual space, the virtual space reference marker representing the object or the image including the object, and storing the generated virtual space reference marker in the real space Interlocking with the fiducial marker And generating a virtual space reference marker. 23. The method of claim 22, The step (b) (b) generating a virtual space model having the same size as the real space; (bb) inputting a corresponding object corresponding to the object included in the image of the real space into the generated virtual space; And (bc) generating the virtual space reference marker to be connected to the input corresponding object And generating a virtual space reference marker. 23. The method of claim 22, The step (a) (aa) pointing to the real space; (ab) acquiring an image of the pointed reality space; (ac) acquiring context information including location information and azimuth information of an object included in the acquired image; And (ad) recognizing the actual space reference marker by considering the obtained image and the obtained context information, and estimating a posture And generating a virtual space reference marker. An operation method of a marker management system for managing a marker for expression of an augmented reality, (a) recognizing an object included in an image of a pointed physical space; (b) extracting a virtual space reference marker having a maximum confidence value for the recognized object; (c) recognizing the actual space reference marker included in the image and estimating the posture; And (d) overlaying information on the extracted virtual space fiducial markers on the recognized real space fiducial markers And generating a virtual space reference marker. 26. The method of claim 25, In the step (b), a virtual space reference marker having the largest number of objects in the representative image is extracted based on the coordinate information of the recognized objects with the maximum virtual space reference marker having the maximum confidence value, Wherein the step (d) comprises performing the overlay after matching the coordinate information of the extracted virtual space reference marker with the coordinate information of the recognized real space reference marker. A computer-readable recording medium, 26. A recording medium on which a program for implementing the method according to any one of claims 19 to 26 is recorded.

Images