JP2000250942A - Book information search space construction / search apparatus and computer-readable recording medium recording book information search space construction / search program - Google Patents

Abstract

(57) [Summary] [Problem] To construct a classification system in a three-dimensional space from an object in a three-dimensional space, which is input data, and a classification system of a book, and to search for book information. SOLUTION: A CG for constructing a three-dimensional search space by associating a book classification system with a computer graphics object or an image object in a three-dimensional space by data input by data input means 11.
Automatic search space construction means 12, partial anchor construction means 13 for constructing partial anchors on objects in the three-dimensional search space, display means 51 for displaying the three-dimensional search space constructed by the search space construction part 10, and mouse instruction Means 52
3D search means 53 for searching for information on the link destination of the anchor selected by the point instruction by the user, and the book information using the book classification system information associated with the link destination of the anchor selected by the point instruction. A book search means 55 for searching is provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention constructs a three-dimensional search space for searching target book information while searching for related information in a virtual three-dimensional search space, The present invention relates to a book information search space construction / search apparatus for searching a book, and a computer-readable recording medium storing a book information search space construction / search program.

[0002]

2. Description of the Related Art As a tool for handling three-dimensional information on a computer, for example, VRML (Virtual Reale
Cyber Modeler (trademark of Toyo Information Systems Co., Ltd.), which is an authoring tool, and Community Place (trademark of Sony Corporation), which is a VRML browser, are known.

[0003] Cyber Walker arranges a virtual object in a virtual three-dimensional space on a computer, sets a color, lighting, and the like, constructs a virtual space, and displays the virtual space on a display screen. In Cyber Walker, it is possible to construct a three-dimensional search interface by adding relevant information to a virtual object.

[0004] In the Community Place, when an operator operates a cursor on a screen using a pointing device such as a mouse in a virtual three-dimensional space to walk through the virtual space or designate a virtual object on the screen. In addition, related information such as a homepage on the Internet associated with the virtual object can be searched.

FIG. 31 shows a configuration of a conventional three-dimensional information search apparatus for searching for three-dimensional information in a virtual space. The three-dimensional information search device includes a search space construction unit 200,
A search unit 300, an object database 101,
An anchor / link attribute database 102 storing data of anchors, links, and attributes. The search space construction unit 200 includes a data input unit 201 for inputting data necessary for constructing a three-dimensional search space, a three-dimensional search space / manual construction unit 204 including an anchor construction unit 202 and a link information setting unit 203. The search unit 300 includes a mouse instruction unit 301 and a display unit 30
2 and a three-dimensional search means 303.
Objects handled by the search space construction unit 200 and the search unit 300 are both based on VRML data.

Next, the operation will be described. In the search space construction unit 200, a three-dimensional search space / manual construction means 204
To construct a three-dimensional space, and the anchor construction means 202
A link can be constructed by designating an object to be an anchor by using and specifying a URL of a link destination by the link information setting means 203. All information on anchors and link destinations handled here is stored in VRML data which is an object.

The search unit 300 performs a search while walking through a three-dimensional space using VRML data having information such as anchors and links. When an object in VRML is displayed in a three-dimensional space by the display means 302 and an anchor is designated by the mouse instructing means 301, the three-dimensional search means 303 causes the
RL (Uniform Resource Locat
or), and displays the link destination resource by displaying means 302
To be displayed.

[0008]

However, when searching for book information using the above-described conventional three-dimensional information search apparatus, it is necessary to prepare objects corresponding to each item of the book classification system. In addition, it is necessary to construct a partial anchor in the object in accordance with the hierarchical relationship of the classification system, and to associate (link) to a lower item from the object, and it takes time to construct a search space.

In the above-mentioned conventional three-dimensional information retrieval apparatus, if there is no computer graphics (CG) object corresponding to the item of the book classification system, the three-dimensional information retrieval apparatus is not used.
It is necessary to construct a CG object and an anchor in the dimensional modeler, and this also takes time to construct a search space. Also, when the number of polygons of the CG object is large and the data amount is large, it takes time to read and display data on the three-dimensional space, and it takes time to move the viewpoint, so that natural walkthrough is difficult. There was a problem.

In the conventional three-dimensional information retrieval apparatus, it is necessary to associate all CG objects and anchors in consideration of the hierarchical relationship of the book classification system.
Construction of the search space was troublesome, and it was necessary to confirm whether or not each item in the classification system was missing.

Further, the association between the CG object and the anchor in the conventional three-dimensional information retrieval apparatus is defined by a static link, and when the object or the anchor is redefined, the link information is also redefined. Since the search space must be changed, it takes time to change the search space, and it is difficult to manage each data independently.

Further, in the conventional three-dimensional information retrieval device,
URL as the link destination defined for the dimensional object
Although a hyperlink can be executed to display a related homepage, there is a problem that book information cannot be searched directly from anchor information.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem. In the construction of a book information search space using a three-dimensional search space, an object on a three-dimensional space which is input data and a book A classification system in a three-dimensional space can be constructed from the classification system. For each item of the classification system,
Even if all the objects are not prepared, the corresponding objects and anchors can be constructed to define the hierarchical relationship, and the classification system can be constructed on the three-dimensional search space to search for related information while the final information is being searched. A system for searching for book information has been constructed to enable searching for the desired book while viewing visual three-dimensional information. In addition to being able to express not only CG objects but also video data, it is possible to replace CG objects with many polygons. A book information search space construction / search apparatus and a book information search space construction / search program which can construct a three-dimensional search space by using data as well as independently manage object data by dynamically setting links are recorded. It is intended to obtain a computer-readable recording medium.

[0014]

In order to achieve the above object, a book information search space construction / search apparatus according to the present invention stores related information in a virtual three-dimensional search space constructed on a computer. A book information search space construction / search apparatus for constructing a search space for searching for target book information while searching, and for searching for book information, comprising a three-dimensional search space for book information search. Data input means for inputting the book classification system required for construction, CG object and image object data in the three-dimensional space, and the book classification system and the three-dimensional space based on the data input by the data input means. Automatic search space construction means for constructing a three-dimensional search space by associating a CG object or an image object with a part of the object in the three-dimensional search space A search space construction unit configured by partial anchor construction means for constructing an anchor; a display means for displaying the three-dimensional search space constructed by the search space construction unit; a pointing device for pointing to the three-dimensional search space; Using three-dimensional search means for searching for information on the link destination of the anchor selected by the pointing instruction of the pointing device and the classification system information of the book associated with the link destination of the anchor selected by the pointing instruction of the pointing device And a search unit including book search means for searching for book information.

Construction of a book information search space according to the following invention
A search device, wherein the search space automatic construction means associates each item of the book classification system with the input object, and determines a necessary construction and setting according to the state of association; If there is no object corresponding to the item of the book classification system, and if there are multiple objects in the lower item, an object automatic construction unit that automatically generates and associates the object of the upper item from the information , An automatic anchor construction unit that automatically constructs an anchor required as a link source for associating a higher-order item with a lower-order item in a book classification system, and information on the book classification system as anchor attribute information And an attribute information automatic setting unit that automatically sets the attribute information.

Construction of book information search space according to the following invention
The search device builds a search space for searching for desired book information while searching for related information in a virtual three-dimensional search space built on a computer, and searches for book information. A book information search space construction / search apparatus for inputting data of a book classification system and CG objects and image objects in a three-dimensional space necessary for constructing a three-dimensional search space for book information search. Search space automatic construction means for constructing a three-dimensional search space by associating a book classification system with a CG object, image object or video frame object in a three-dimensional space based on the data inputted by the input means and the data input means. And a partial anchor construction means for constructing a partial anchor on an object in the three-dimensional search space A cable space construction unit, display means for displaying the three-dimensional search space constructed by the search space construction unit, and video display for performing processing for displaying a video frame on a virtual panel, which is a single polygon in the three-dimensional search space. A processing device, a pointing device that points to the three-dimensional search space, a three-dimensional search device that searches for information on a link destination of the anchor selected by the pointing device of the pointing device, and an anchor selected by the pointing device of the pointing device. A book search unit configured to search for book information using book classification system information associated with the link destination.

Construction of a book information search space according to the following invention
A search device, wherein the search space automatic construction device associates each item of the book classification system with the input object, and determines a necessary construction and setting in accordance with the state of association; Using a CG object of a lower item or a CG object of its own, a rendering image of the CG object from multiple viewpoints is taken,
A video frame, which is a set of rendering images for each viewpoint, is automatically generated and registered in a database so that viewpoints can be moved in a plurality of directions on a three-dimensional search space instead of a CG object, and the object is used as an item. A video frame automatic construction unit for associating, and a video frame / anchor automatic construction unit for automatically constructing, for a video frame object, an anchor required as a link source for associating an upper item to a lower item. And an attribute information automatic setting device for automatically setting information as attribute information of an anchor.

Construction of book information retrieval space according to the following invention
A search device, wherein the video display processing means determines a viewpoint of a searcher; a video frame selection unit that selects a video frame from a database in accordance with the position of the viewpoint; and a video frame in accordance with the position of the viewpoint. A virtual panel / position setting unit that determines the position of a virtual panel, which is a single polygon in a three-dimensional search space for mapping a video, and a video mapping unit that maps a selected video frame to the virtual panel. Is what it is.

Construction of a book information search space according to the following invention
In the search device, the video display processing means may further obtain, at a viewpoint position where a video frame is not prepared, interpolation data from data of surrounding video frames already prepared and use the data as a selected frame. It includes an interpolation unit.

Construction of book information search space according to the following invention
The search device builds a search space for searching for desired book information while searching for related information in a virtual three-dimensional search space built on a computer, and searches for book information. A book information search space construction / search apparatus for inputting data of a book classification system and CG objects and image objects in a three-dimensional space necessary for constructing a three-dimensional search space for book information search. Search space automatic construction means for constructing a three-dimensional search space by associating a book classification system with a CG object, image object or moving video object on the three-dimensional space based on the data inputted by the input means and the data input means. And a partial anchor construction means for constructing a partial anchor on an object in the three-dimensional search space Video display processing for displaying a moving video on a virtual panel, which is a single polygon in the three-dimensional search space, and a display means for displaying the three-dimensional search space constructed by the search space construction part, Means, a pointing device that points to the three-dimensional search space, a three-dimensional search means that searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device, and a link of the anchor selected by the pointing instruction of the pointing device. A book search unit configured to search for book information using the book classification system information associated with the book.

Construction of a book information search space according to the following invention
A search device, wherein the search space automatic construction device associates each item of the book classification system with the input object, and determines a necessary construction and setting in accordance with the state of association; Using the CG object of the lower order item or the CG object of its own, a moving video that captures the state of viewpoint movement from a plurality of viewpoints of the CG object is automatically generated, and a plurality of moving videos are generated in the three-dimensional search space instead of the CG object. The moving video automatic construction unit registers the object in the database so that the viewpoint can be moved in the direction and associates the object with the item, and it is necessary as a link source to link the upper item to the lower item. A moving video anchor automatic construction unit for automatically constructing an Those that contain the attribute information automatic setting device automatically set as the attribute information over.

Construction of a book information search space according to the following invention
The search device, the video display processing means, a viewpoint determination unit that determines the viewpoint of the searcher, a moving video selection unit that selects a moving video from the database according to the position of the viewpoint,
A virtual panel / position setting unit that determines a position of a virtual panel, which is a single polygon in a three-dimensional search space for mapping a moving video according to a viewpoint position, and maps the selected moving video to the virtual panel. And a video mapping unit.

Construction of a book information search space according to the following invention
In the search device, the video display processing means may further include, at a viewpoint position where a moving video is not prepared, obtaining by interpolation processing data of surrounding moving videos already prepared and using the data as a selected moving video. It includes a video interpolation unit.

Construction of a book information search space according to the next invention
In the retrieval apparatus, the retrieval unit has attribute information retrieval means for retrieving related information from the attribute information of the anchor selected by the pointing instruction of the pointing device.

In order to achieve the above object, a computer-readable recording medium storing a book information search space construction / search program according to the present invention is provided on a virtual three-dimensional search space constructed on a computer. A computer-readable recording medium storing a program for retrieving book information, constructing a search space for retrieving target book information while searching for related information of A search space construction for constructing a three-dimensional search space by associating a book classification system with a CG object or an image object on the three-dimensional space based on data input by the means, and constructing a partial anchor for the object in the three-dimensional search space And displaying the three-dimensional search space constructed by executing the search space construction procedure. Searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device, and searches for book information using the classification system information of the book associated with the link destination of the anchor selected by the pointing instruction of the pointing device And a search procedure to be performed.

Construction of a book information search space according to the following invention
The computer-readable recording medium on which the search program is recorded, wherein the search space construction procedure associates each item of the book classification system with the input object and performs the necessary construction and setting according to the association status. Judgment, if there is no object corresponding to the item of the book classification system, and if there are multiple objects in the lower item, automatically generate the object of the upper item from the information and make a correspondence, Automatically constructs an anchor required as a link source for associating higher-order items with lower-level items in the book classification system, and automatically sets information in the book classification system as anchor attribute information Things.

Construction of a book information search space according to the following invention
A computer-readable recording medium on which a search program is recorded is a search space for searching for target book information while searching for related information in a virtual three-dimensional search space built on a computer. A computer-readable recording medium storing a program for retrieving book information and a book classification system and a CG object, an image object, or a video frame in a three-dimensional space based on data input by data input means. A three-dimensional search space is constructed by associating the object with the search space, and a search space construction procedure for constructing a partial anchor for an object in the three-dimensional search space, and the three-dimensional search space constructed by executing the search space construction procedure are displayed. Then, a video frame is a virtual panel that is a single polygon in the three-dimensional search space. To display the information of the link destination of the anchor selected by the pointing instruction of the pointing device, and retrieves the classification system information of the book associated with the link destination of the anchor selected by the pointing instruction of the pointing device. And a retrieval procedure for retrieving book information by using the information.

Construction of a book information search space according to the following invention
The computer-readable recording medium on which the search program is recorded, wherein the search space construction procedure associates each item of the book classification system with the input object and performs the necessary construction and setting according to the association status. Judgment, taking a rendering image of the CG object from a plurality of viewpoints using the CG object of the lower item or the CG object of itself, automatically generating a video frame as a set of rendering images for each viewpoint, and In order to register viewpoints in multiple directions on the three-dimensional search space instead of objects in the database, to associate the objects with items, and to associate items from higher to lower items, The anchor required as a link source is automatically set for the video frame object. To build, in which automatically sets the information as the attribute information of the anchor.

Construction of book information search space according to the following invention
Computer-readable recording medium recording a search program, the search procedure, determine the viewpoint of the searcher,
A video frame is selected from the database in accordance with the position of the viewpoint, and the position of a virtual panel, which is a single polygon, in a three-dimensional search space for mapping the video frame in accordance with the position of the viewpoint is determined. It maps video frames to virtual panels.

Construction of book information search space according to the following invention
The computer-readable recording medium on which the search program is recorded, wherein the search procedure further obtains the data obtained by interpolation processing from data of surrounding video frames already prepared at a viewpoint position where video frames are not prepared. Is the selected frame.

Construction of book information search space according to the following invention
A computer-readable recording medium on which a search program is recorded is a search space for searching for target book information while searching for related information in a virtual three-dimensional search space built on a computer. A computer-readable recording medium on which a program for retrieving book information is constructed, wherein a book classification system and a CG object, an image object or a moving video on a three-dimensional space are recorded according to data inputted by data input means. A three-dimensional search space is constructed by associating the object with the search space, and a search space construction procedure for constructing a partial anchor for an object in the three-dimensional search space, and the three-dimensional search space constructed by executing the search space construction procedure are displayed. And
Performs processing to display the moving video on a virtual panel, which is a single polygon in the three-dimensional search space, searches for the link destination information of the anchor selected by the pointing instruction of the pointing device, and selects it by the pointing instruction of the pointing device And a search procedure for searching for book information using the book classification system information associated with the link destination of the specified anchor.

Construction of book information search space according to the following invention
The computer-readable recording medium on which the search program is recorded, wherein the search space construction procedure associates each item of the book classification system with the input object and performs the necessary construction and setting according to the association status. Judgment is made, and using the CG object of the lower item or the CG object of one's own, a moving video that captures the state of viewpoint movement from a plurality of viewpoints of the CG object is automatically generated, and a three-dimensional search space is used instead of the CG object Register the object in the database so that it can move the viewpoint in multiple directions above, associate the object with the item, and specify the anchor required as a link source to associate the upper item with the lower item. Automatically constructs moving video objects and automatically sets information as anchor attribute information. It is intended to.

Construction of a book information search space according to the following invention
Computer-readable recording medium recording a search program, the search procedure, determine the viewpoint of the searcher,
The moving video is selected from the database according to the position of the viewpoint, and the position of the virtual panel, which is a single polygon, in the three-dimensional search space for mapping the moving video according to the position of the viewpoint is determined. The moving video is mapped to the virtual panel.

Construction of book information search space according to the following invention
The computer-readable recording medium on which the search program is recorded, wherein the search procedure further obtains, at a viewpoint position where the moving video is not prepared, an interpolation process from data of the surrounding moving video that is already prepared. Is the selected moving video.

Construction of a book information search space according to the following invention
In the computer-readable recording medium on which the search program is recorded, the search procedure searches the attribute information of the anchor selected by the pointing instruction of the pointing device for related information.

[0036]

BEST MODE FOR CARRYING OUT THE INVENTION A book information search space construction / search apparatus and a book information search space construction / search apparatus according to the present invention will be described below.
An embodiment of a computer-readable recording medium on which a search program is recorded will be described in detail with reference to the accompanying drawings.

Embodiment 1 FIG. 1 shows a first embodiment of a book information search space construction / search apparatus according to the present invention. The book information search space construction / search device includes a search space construction unit 10 and a search unit 50.

The search space construction unit 10 is a data input means for inputting a book classification system necessary for constructing a three-dimensional search space for book information retrieval and data of CG objects and image objects in the three-dimensional space. CG search space automatic construction means 12 for automatically associating a book classification system with a CG object or image object in a three-dimensional space from data input by the data input means 11 and constructing a three-dimensional search space And a partial anchor construction unit 13 for constructing a partial anchor on a CG object or an image object in the three-dimensional search space.

The search unit 50 searches for a target book by searching for related information in the three-dimensional search space constructed by the search space construction unit 10. The three-dimensional search space constructed by the search space construction unit 10 Display means 51 for displaying a CG object and an image object on a screen, a mouse instruction means 52 for designating a search target in a three-dimensional search space, and a link destination of the search target (anchor) selected by the instruction of the mouse instruction means 52 A three-dimensional search unit 53 for searching for three-dimensional information, an attribute information search unit 54 for searching for related information from the search target attribute information selected by the instruction of the mouse instruction unit 52, and a selection by the instruction of the mouse instruction unit 52 Book search that searches linked book information using the classification system information of the book associated with the link of the searched search target (anchor) And a stage 55, and displays a result of searching by their search means on the display means 51.

Information on CG objects and image objects displayed in the three-dimensional search space is stored in the object database 101, and classification system information indicating the item names and hierarchical relationships of each item of the book classification system is stored in the classification system information database 100. , Book information is the book information database 10
3 can be stored in advance.

Next, the operation will be described. First, in the search space construction section 10, the data input means 11 inputs the item names of the respective items of the book classification system and the classification system information indicating the hierarchical relationship from the classification system information database 100, and displays them in the three-dimensional search space. Information of an object to be input is input from the object database 101. As an example, it is assumed that the input object is “heart, lung, stomach, intestine, liver” in the book classification items illustrated in FIG. Of those objects, the parts corresponding to the subitems of “stomach, intestine, liver” are “cardia, fundus, pylorus, antrum, stomach, stomach wall”, “large intestine, small intestine”, “left lobe, right lobe” For, it is assumed that the partial anchor has already been constructed.

The constructor associates each item of the book classification system of the input data with the information of the object by the automatic CG search space construction means 12. If an object is not prepared for an item, an object and an anchor are automatically constructed using the information of the already associated object. For example, when there is an object in a lower item, in the upper item, a plurality of lower objects are simultaneously displayed to express the concept of the upper item.

When it is necessary for the builder to construct a partial anchor, the construction is supported by the partial anchor construction means 13. In the partial anchor construction means 13, for example, 3
When the dimensional modeler designates two points in the three-dimensional space, the polygon included in the rectangular parallelepiped (the z-direction is infinite) composed of the two points is determined and defined as a new partial anchor.

The link relation of the objects in the three-dimensional space is defined from the information on the hierarchical relation of the classification system. For example,
When linking a stomach object directly from an upper-level abdominal object, a link is defined from the “abdominal” anchor to the “stomach” anchor. The link information realizes a hierarchical relationship of the classification system in the three-dimensional search space. Anchor, link, and attribute data are stored in the anchor / link attribute database 102.

The three-dimensional search space including the CG object and the image object constructed by the search space construction unit 10 is displayed on the display means 51 of the search unit 50, and the searcher uses the mouse instruction means 52 to search for the anchor related to the book. To search for more related information. If the upper item is represented by a plurality of lower items,
By causing the searcher to select a copy, it is possible to trace the hierarchy to lower items.

For the three-dimensional related information, the link destination of the anchor selected by the searcher with the mouse pointing means 52 is searched by the three-dimensional search means 53, and the search result is displayed on the display means 51. For example, if you select the abdomen,
Search for "stomach" at the link destination. When an image file is set as an object, an anchor keyword or the like is set as in the case of the CG object, and the display unit 51 maps and displays the virtual panel, which is a single polygon of the CG object. I do.

The attribute information searching means 54 searches the attribute information of the anchor selected by the searcher using the mouse pointing means 52 for related information. The result of selecting the anchor related to the book to be searched by the mouse pointing means 52 is also passed to the book search means 55, and the book information database 10
3 is searched for the book information stored in 3 and the search result is displayed. For example, when the anchor of "stomach" is selected on the display means 51, the book search means 55 searches the book information using the anchor keyword "stomach" as a search keyword and displays the result. I do.

As described above, even if a CG object or an anchor corresponding to each item of the classification system is not prepared, a classification system in a three-dimensional space is constructed while maintaining consistency with the hierarchical relationship of the classification system. Since it is possible, the trouble of the builder can be reduced, and a book search interface compatible with the classification system can be constructed.

Further, by selecting an anchor in the three-dimensional search space and searching for the related three-dimensional information and book information, it is possible to narrow down the target information visually, so that only a large amount of book information can be obtained. In this case, effective narrowing down is possible as compared with the conventional method shown in FIG.

Next, the details of the CG search space automatic construction means 12 will be described with reference to FIG. A CG search space automatic construction means 12 for associating each item of the book classification system with the input object and determining a necessary construction, setting, and the like according to a correspondence situation; If there is no object corresponding to the item of the classification system of the above, and if there are a plurality of CG objects in the lower item, the CG object of the upper item is automatically generated from the information and automatically associated CG object construction Part 2
1 and a CG anchor automatic construction unit 22 for automatically constructing an anchor required as a link source for associating a higher order item to a lower order item in the book classification system.
And an attribute information automatic setting unit 23 for automatically setting information on a book classification system as attribute information of an anchor.

The information on the book classification system set by the attribute information automatic setting unit 23 as the attribute information of the anchor includes information (concept keyword) relating to the association from the upper item to the lower item in the book classification system, and narrowing down. There are keywords and so on.

The processing outline of the correspondence setting unit 20 will be described with reference to FIG. In the interface (screen display) of the correspondence as exemplified in FIG. 8, the file name of the object such as “Stomach.wrl” is input in the input field of the corresponding object, and the corresponding item is classified into the right category. Select from the diagram of the system.

The correspondence setting section 20 performs a correspondence check process and, based on the input information for registering in the database among the input data, converts data such as an object file name, a corresponding item name, and an anchor type into the object database. The object is registered in the database by an object database (DB) registration process, and an item name corresponding to the anchor keyword of the object is set by an anchor keyword (KW) setting process. Examples of the type of anchor include an anchor corresponding to an object, an anchor corresponding to a synthesized object obtained by synthesizing a plurality of objects, a partial anchor, and the like, and these can be set.

All the prepared objects are associated with each other, and a menu button of "automatic construction" in FIG. 8 is pressed to determine items to which no objects are associated and objects for which an anchor is not prepared. Instruct the user to do so. When the correspondence of the prepared objects is completed, in the correspondence check processing of the correspondence setting unit 20, it is checked whether there is a corresponding anchor in each item of the classification system, and the uncorresponding item is determined by the CG object. It is passed to the automatic construction unit 21.

The processing outline of the CG object automatic construction unit 21 will be described with reference to FIG. The CG object automatic construction unit 21 first performs a lower item check process, and checks whether or not there is a corresponding object of the lower item.
If a plurality of objects exist at a lower level, these synthesized objects are generated by object synthesis / database (DB) registration processing, registered in the database, and the corresponding item names are set by anchor keyword (KW) setting processing. .

For example, if there are heart and lung objects but no chest object, the heart and lung objects are combined and the chest item (heart and lung, which is the upper item, is displayed simultaneously at an individual position). Is automatically generated and registered in the database as a chest item. At this time, the anchor keyword of the composite object is the chest. To build the composite file,
For example, in the case of VRML, an object corresponding to a lower item is read using an Inline node.

Next, the CG anchor automatic construction unit 22 automatically constructs a partial anchor corresponding to a lower-order item for the item for which the composite object is created. The processing outline of the CG anchor automatic construction unit 22 will be described with reference to FIG.
An automatic anchor construction / database (DB) registration process is executed, and the original file combined with the combined file is registered in the database as an anchor. Then, the subordinate item names are registered in the database as anchor keywords by the anchor keyword (KW) setting process.

FIG. 9 shows an example of a database in which anchor information is registered. In FIG. 9, "composite anchor" is the anchor of the composite object, "Obj
ect anchor ”is an anchor corresponding to the object,
“Partial anchor” is an anchor corresponding to a part of an object.

In the correspondence check by the correspondence relation setting unit 20, if it is found that the anchor is not set for the lower item, although it is not a composite file, in the example, the right atrium and the left If an anchor corresponding to "atria" is not prepared, the partial anchor construction means 13 causes the builder to define a partial anchor.

In the partial anchor construction means 13, for example, when the three-dimensional modeler designates a certain two points in the three-dimensional space, a rectangular parallelepiped formed by the two points (z direction is infinite)
Is determined and defined as a new partial anchor. With respect to the defined partial anchor, in the anchor keyword setting process, lower item names “right atrium” and “left atrium” are registered in the database as respective anchor keywords.

When the anchor type is a partial anchor, the attribute keyword (K
W) Through the setting process, “heart” is set as a conceptual keyword in each partial anchor of “right atrium” and “left atrium”. Finally, the attribute information automatic setting unit 23 checks the attribute information for the anchors of all the corresponding items, and sets the information if it is not set.

FIG. 6 shows an outline of the processing of the attribute information automatic setting unit 23.
This will be described with reference to FIG. The attribute information automatic setting unit 23 checks an anchor for which no attribute information has been set by the attribute information check processing. If the anchor keyword which is the attribute information is not set, an inquiry is made to the builder to set it.

If the concept keyword is not set, the type of the upper anchor corresponding to the link source is checked by the anchor type check process, and the concept keyword matched to the upper anchor type is set by the concept keyword (KW) setting process. (See FIG. 9). If the object of the upper item is a link source anchor, the anchor keyword of the object is set as a concept keyword. If the partial anchor of the object of the upper item is a link source anchor, the anchor of the partial anchor is set. Set keywords as concept keywords. As a result, the information of association from the upper item to the lower item in the book classification system is automatically set as the attribute information of the anchor.

The interface shown in FIG. 8 is an example, and does not need to be in this format. For example, menu buttons such as "association" and "automatic construction" may be realized in other formats.

As described above, even if a CG object corresponding to each item of the classification system or an anchor is not prepared, an object of a higher order item is automatically generated from information of a lower order item, and an association of items is established. It constructs the anchors necessary for this, and also defines the link information that indicates the hierarchical relationship. Therefore, it is not necessary to prepare all the CG objects and anchors and construct all the links, so that the labor can be reduced.

In addition, a plurality of lower objects
Since we set conceptual keywords for lower-level objects,
One-to-N (N> 1) link definition and link execution are possible, and the link relationship is defined not by the ID or address of the anchor but by the concept information, so that not only the determined database information but also various Database information, WEB
It is also possible to execute a link like a keyword search such as accessing information on a site and searching for related information.

Since the concept information of the object is defined, the anchor of the link source is redefined, and if the concept information does not change even if the ID is changed, it is necessary to change the concept keyword of the linked object. Absent. That is, even if only the link source changes, the search can be performed as it is.

As described above, the link relationship is not statically linked, but attribute information is defined for each anchor, and a link destination is dynamically searched for at the time of search. Is changed, the influence on other objects and the like is small, and independent management of data becomes possible.

Further, the attribute information automatic setting section 23 sets a narrowing keyword (see FIG. 9) as the anchor attribute information by a narrowing keyword setting process.

For example, when the classification system items under the concept of “structure” are put together and automatically constructed as a three-dimensional search interface, the concept of “structure” is
Automatically set to narrowing keywords of all items.

As a result, when a plurality of anchors having the anchor keyword “stomach” are obtained as a search result, when the searcher inputs the search purpose “structure”, for example, “sickness”, “function” An anchor having other narrowing keywords such as is not extracted, and a narrowing search that displays only a search result of “structure” becomes possible.

Embodiment 2 FIG. 10 shows a second embodiment of the book information search space construction / search apparatus according to the present invention. In FIG. 10, components that are the same as or the same as those shown in FIG. 1 are given the same reference numerals as those shown in FIG. 1, and descriptions thereof are omitted. The book information search space construction / search device has a search space construction unit 30 and a search unit 60.

The search space construction unit 30 includes a data input unit 11 for inputting a book classification system necessary for constructing a three-dimensional search space for book information search, a CG object and an image object in the three-dimensional space, and A three-dimensional search space by automatically associating a book classification system with a CG object, an image object, or an object based on a video frame (a collection of still image data) in a three-dimensional space from data input by the data input means 11. CG / video search space automatic construction means 14 for constructing a partial anchor, and a partial anchor construction means 13 for constructing a partial anchor for an object in the three-dimensional search space.

The search section 60 searches for a target book by searching for related information in the three-dimensional search space constructed by the search space construction section 30. The three-dimensional search space constructed by the search space construction section 30 A display means 51 for displaying a CG object, an image object, and an object based on a video frame; a video display processing means 56 for performing processing for displaying video data as an object on the display means 51 in a three-dimensional search space; A mouse instructing means 52 for instructing a search target in the dimensional search space; and a CG / video search means (three-dimensional) for searching for link destination information (related information) of the search object (anchor) selected by the instruction of the mouse instructing means 52 Search means) 57 and related information based on the attribute information of the search target (anchor) selected by the instruction of the mouse instruction means 52 Attribute information retrieval means 54 to search for broadcast
And a book search unit 55 that searches for linked book information by using book classification system information associated with the link of the search target (anchor) selected by the instruction of the mouse instruction unit 52, The results searched by the search means are displayed on the display means 51.

Next, the operation will be described. First, in the search space construction section 30, the data input means 11 inputs the item name of each item of the book classification system and the classification system information indicating the hierarchical relationship from the classification system information database 100, and displays it in the three-dimensional space. Object information is input from the object database 101.

The CG / video search space automatic construction means 14 allows the builder to associate each item of the book classification system of the input data with the object information. If an object is not prepared for an item, an object and an anchor are automatically constructed using the information of the already associated object.

CG / video search space automatic construction means 14
Indicates that the CG object has a large number of polygons and 3
It is slow to display and walk through in dimensional space,
In the case of unnaturalness, a three-dimensional CG is displayed in a three-dimensional space, a hard copy is taken, video data (still image data) is automatically generated, and a video frame composed of a group of video data is used as an object. I do. For example, a video frame in which the viewpoint is moved once in the horizontal direction around the object based on the three-dimensional CG is automatically generated and stored instead of the object.

By adding a video in all directions, such as a video frame moved one round in the vertical direction and a video frame moved one round in the diagonal direction, reference from various viewpoints becomes possible. . The data format of these video frames is saved as an image one by one,
When displaying, select the best image to display,
Alternatively, it is also possible to store the video data in a certain unit, select an optimum video according to the viewpoint direction, and select a video frame from the optimum video.

When the construction of the partial anchor by the builder is necessary, the construction is supported by the partial anchor construction means 13. In the partial anchor construction means 13, for example, 3
When the dimensional modeler designates two points in the three-dimensional space, the polygon included in the rectangular parallelepiped (the z-direction is infinite) composed of the two points is determined and defined as a new partial anchor.

The three-dimensional search space constructed by the search space construction unit 30 is displayed on the display means 51 of the search unit 60, and the searcher selects an anchor related to the book to be searched by the mouse instruction means 52, and updates the search. Search for relevant information.
When the object displayed on the display means 51 is a video frame, the video data is displayed on the video display processing means 5.
6 is processed so as to be displayed on a virtual panel which is a single polygon in a three-dimensional space, and this virtual panel can be used as one anchor to link to lower-level information. 3
As for the dimension-related information, the search target searches for the link destination of the anchor selected by the mouse instruction means 52 by the CG / video search means 57 and displays the search result.
To display.

As described above, a classification system in a three-dimensional search space can be constructed using not only CG objects and images but also video data. If the number of polygons of the CG object is large, it is substituted with video data. By doing so, the speed of the walk-through can be increased.

In the above description, the automatic CG / video search space construction means 14 automatically creates a video frame from a CG object. However, a real object is photographed by a digital still camera, a video camera, or the like. It is also possible to input this image data, process this image data by the video display processing means 56 and display it on the display means 51 in the same manner, and this method can be used also for actual video data. This is effective when it is desired to use real data that cannot be obtained as a three-dimensional CG model.

Next, details of the CG / video search space automatic construction means 14 will be described with reference to FIG. The automatic CG / video search space construction means 14 includes a correspondence setting unit 20,
In addition to the CG object automatic construction unit 21, the CG anchor automatic construction unit 22, and the attribute information automatic setting unit 23, it has a video frame automatic construction unit 24 and a video frame anchor automatic construction unit 25.

The video frame automatic construction unit 24 captures a rendering image of the CG object from all viewpoints using the CG object of the lower item or the CG object of its own, and a video frame which is a set of rendering images for each viewpoint. Is automatically generated, registered in the database so that the viewpoint can be moved in any direction in the three-dimensional space instead of the CG object, and the object is associated with the item.

Video frame anchor automatic construction unit 25
Automatically establishes an anchor required as a link source for an object based on a video frame created by the video frame automatic construction unit 24 in order to associate an upper item with a lower item. Similar to the first embodiment, the correspondence setting unit 20 associates a classification system with an object based on the data described above. When the correspondence of the prepared objects is completed, in the correspondence check processing by the correspondence setting unit 20, it is checked whether or not there is a corresponding anchor in each item of the classification system, and the uncorrelated item is selected. Is passed to the automatic video frame construction unit 24.

The processing outline of the automatic video frame construction unit 24 will be described with reference to FIG. The video frame automatic construction unit 24 first performs lower-level item check processing, checks whether or not there is a corresponding object of the lower-level item, and if a plurality of objects exist at lower levels, the synthesized objects are subjected to object synthesis processing. Generate and shoot rendered images of the composite object from various viewpoints by video frame shooting processing.
In the video frame database (DB) registration processing, the video frame is registered in the database. Note that the video frame automatic construction unit 24 has a default grid point interval for capturing video frames, but the builder can change the grid point interval.

For example, if there are heart and lung objects but no chest object, the heart and lung objects are combined to generate a chest object, and rendered images are taken from various viewpoints, and the chest image is taken. Register in the database as an item. The rendered image at each viewpoint is called a video frame. The captured video frame is registered in a video frame database (DB) registration process so that the position of the viewpoint can be known.

Next, the processing outline of the video frame anchor automatic construction section 25 will be described with reference to FIG. The video frame anchor automatic construction unit 25 registers the video frames constructed by the video frame automatic construction unit 24 as anchors in the database by the video frame anchor database (DB) registration processing, and sets the anchor keyword (KW). An anchor keyword is set by processing.

For example, a video frame constructed with respect to the chest of one CG object is photographed for several viewpoints, but has one anchor and all anchor keywords are “chest”. Therefore, as in the case of other CG objects, the anchor database may be registered as one anchor, and the actual frame data may be managed in another database. FIG. 14 shows an example of a database in which anchor information is registered. In FIG. 14, "VF anchor" is a video frame anchor. In the attribute information automatic setting 23, when the anchor type is a video frame, the anchor keyword of the video frame is set as the conceptual keyword of the lower item of the video frame.

Next, the details of the video display processing means 56 will be described with reference to FIG. Video display processing means 56
Is a viewpoint determination unit 70 that determines the viewpoint of the searcher, a video frame selection unit 71 that selects a video frame from the database according to the position of the viewpoint, and a three-dimensional mapping unit that maps the video frame according to the position of the viewpoint. A virtual panel / position setting unit 72 for determining the position of a virtual panel, which is a single polygon in the space; and a video mapping unit 7 for mapping a selected video frame to the virtual panel
And 3.

The video display processing means 56 first determines the viewpoint of the searcher by the viewpoint determination section 70 and constructs and constructs the video frame by the video frame automatic construction section 24 in the video frame selection section 71.
From the registered video frames, a video frame that matches the viewpoint of the searcher is selected, and a virtual panel / position setting unit 7 is selected.
In FIG. 2, a virtual panel is set in accordance with the viewpoint of the searcher as shown in FIG.
3 to perform video mapping.

Specific examples of video frame shooting and selection / display according to the viewpoint will be described. As illustrated in FIG. 17, from the center of the heart of the three-dimensional object,
A video frame, which is a rendering image from a viewpoint with a radius r, is captured. Using polar coordinates as shown in FIG. 18, for example, the viewpoint is moved at intervals of 30 degrees in the horizontal rotation angle θ and 30 degrees in the vertical direction with respect to the radius r, The video frames of the heart at each location are stored.

FIG. 19 is a plan view showing the arrangement of data of video frames shot at rotation angles θ and δ of 30 degrees each. In FIG. 19, one cell is 30 degrees each in the vertical and horizontal directions. Assuming that the viewpoint has moved along the thick line in FIG. 19, the video frame selection unit 71 selects a rendering image for each viewpoint, and in this case, 2
The video frame indicated by the double circle is selected.

In this case, the viewpoint is moved more finely than the grid point interval of 30 degrees. However, when the grid point of the selected video is traced with respect to the actual movement of the viewpoint, It does not follow the viewpoint movement smoothly. To make it possible to select a video frame smoothly with respect to the movement of the viewpoint, the interval between grid points can be reduced. For example, if the interval of 30 degrees is reduced to half of 15 degrees, it is possible to cope with a fine viewpoint movement. But,
For each of the rotation angles θ and δ, if the interval between the lattice points is halved, the data amount is quadrupled.

Therefore, in such an equally-spaced grid, the movement of one diagonal grid is about 1.4 times longer and longer than that of the vertical and horizontal grids. As seen from FIG. 20, the vertical and horizontal movements do not seem to be so large when compared with the diagonal movements, even if there is a slight shift, as shown in FIG.
The video data is taken alternately at the viewpoint position shifted in phase. As a result, it is possible to make the interval between grid points smaller, increase only the video data for the movement in the oblique direction, enable smooth movement, and reduce the data amount to half.

As described above, even if CG objects and anchors corresponding to the respective items of the classification system are not prepared, the information of the upper and lower items can be obtained while maintaining consistency with the hierarchical relationship of the classification system. By creating objects and anchors and defining link information, it is possible to construct a classification system on a three-dimensional search space using CG objects, images, and videos by supporting the work of the builder. As a result, the work of the builder can be reduced, and a book search interface consistent with the classification system can be constructed.

Also, by automatically generating a video frame of a composite object from a CG object, the same expression as that of the CG object is possible in a three-dimensional search space. Relevant information can be searched, and video frames prepared for each viewpoint can be selected and displayed according to the viewpoint of the searcher. Therefore, the searcher can freely walk around the object, similar to a CG object. it can.

Also, since the interval between the grid points at which video frames are photographed can be easily changed, the viewpoint can be moved smoothly in accordance with the high definition required for the object. It is also possible to move the viewpoint in an oblique direction while suppressing the data amount.

As shown in FIG. 21, the video display processing means 56 includes, in addition to the viewpoint determining section 70, the video frame selecting section 71, the virtual panel / position setting section 72, and the video mapping section 73, The video frame interpolating unit 74 may include a frame interpolating unit 74. The video frame interpolating unit 74 interpolates a video frame at a viewpoint position where a video frame is not prepared from data of surrounding (adjacent) video frames that have already been prepared by interpolation processing. Then, the data is used as a selected frame.

In this case, in contrast to obtaining data in an oblique direction and increasing the data to enable smooth video display, when the video moves diagonally, the video frame interpolator is used based on the video data surrounding it. Interpolation data is automatically created by the interpolation processing by 74 to construct and display video data in an oblique direction.

By adding the video frame interpolator 74, it is not necessary to prepare many video frames from the beginning, and it is efficient to create interpolated frames as needed.

Embodiment 3 FIG. FIG. 22 shows another embodiment of the automatic CG / video search space construction means. In FIG. 22, the part corresponding to FIG.
The same reference numerals are given to the reference numerals, and the description is omitted.

CG / video search space automatic construction means 14 '
Has a moving video automatic construction unit 26 and a moving video anchor automatic construction unit 27 in addition to a correspondence setting unit 20, a CG object automatic construction unit 21, a CG anchor automatic construction unit 22, and an attribute information automatic construction unit 23. are doing.

The moving video automatic construction unit 26 automatically generates a moving video (moving image data) obtained by capturing the viewpoint moving from any viewpoint of the CG object by using the CG object of the lower item or the CG object of its own. The generated object is registered in the database so that the viewpoint can be moved in any direction in the three-dimensional space instead of the CG object, and the object is associated with the item. The moving video anchor automatic construction unit 27 automatically constructs an anchor required as a link source for a moving video object in order to associate an upper item with a lower item.

Also in this embodiment, the correspondence setting unit 20
In the same manner as in the first embodiment, the association between the classification system and the object is performed using the data as described above, and when the association of the prepared objects is completed, the association check process by the association setting unit 20 is performed. In each of the items of the classification system, it is checked whether there is a corresponding anchor, and the selected one of the items that are not supported is passed to the mobile video automatic construction unit 26.

The outline of the processing of the mobile video automatic construction unit 26 will be described with reference to FIG. The moving video automatic construction unit 26 first performs a lower-level item check process, checks whether or not there is a corresponding object of the lower-level item, and if a plurality of objects exist at a lower level, the synthesized objects are subjected to the object synthesis process. By generating and moving video shooting processing, videos of the synthesized object moving in eight directions from various viewpoints are shot and registered in a moving video database (DB) registration processing. Note that the moving video automatic construction unit 26 has a default interval between grid points for capturing a moving video, but the builder can change the grid point interval. Also,
In this example, a moving video in eight directions is captured,
It is not always necessary to use eight directions.

For example, if there is a heart and lung object but no chest object, the heart and lung objects are combined to generate a chest object, and moving videos from various viewpoints are photographed. And register it in the database. By continuously reproducing the moving video, the same expression as that of the CG object is possible, which is called a moving video. The captured moving video is registered in a moving video database (DB) registration process so that the position of the viewpoint can be recognized.

Next, the moving video anchor automatic construction unit 2
The processing outline of 7 will be described with reference to FIG. In the moving video anchor automatic construction unit 27, the moving video automatic construction unit 2
In the moving video / anchor database (DB) registration processing, the moving video constructed in step 6 is registered in the database as an anchor, and the anchor keyword (KW) is registered.
In the setting process, an anchor keyword is set.

For example, a moving video constructed with respect to the chest of one CG object is photographed for several viewpoints. However, there is only one anchor, and all anchor keywords are “chest”. Therefore, as in the case of other CG objects, the anchor database may be registered as one anchor, and the actual moving video may be managed in another database. FIG. 25 shows an example of a database in which anchor information is registered.
In FIG. 5, "3DV anchor" is a moving video anchor. In the attribute information automatic setting means, when the anchor type is the moving video, the concept keyword of the lower item of the moving video sets the anchor keyword of the moving video.

Next, the details of the video display processing means in this embodiment will be described with reference to FIG. In addition,
26, parts corresponding to those in FIG. 15 are denoted by the same reference numerals as those in FIG. 15, and description thereof will be omitted.

The video display processing means 56 'includes a viewpoint determining unit 70 for determining the viewpoint of the searcher, a moving video selecting unit 75 for selecting a moving video from the database in accordance with the position of the viewpoint, and a moving video selecting unit 75 in accordance with the position of the viewpoint. A virtual panel / position setting unit 76 that determines the position of a virtual panel, which is a single polygon in a three-dimensional space, for mapping a moving video.
And a video mapping unit 77 that maps the selected moving video to the virtual panel.

The video display processing means 56 'first determines the viewpoint of the searcher in the viewpoint determination section 70, and in the moving video selection section 75, the moving video automatically constructed and registered by the moving video automatic construction section 26. The moving video that matches the searcher's point of view is selected from the
As shown in FIG. 6, a virtual panel is set according to the viewpoint of the searcher, and video mapping is performed by the video mapping unit 77 there.

[0113] Moving video shooting and selection according to the viewpoint
A specific example of the display will be described. As shown in FIG. 17, for example, a moving video, which is a rendering image from the viewpoint of a radius r, is shot from the center of the heart of the three-dimensional object. Using polar coordinates as shown in FIG. 18, for example, the viewpoint is moved at intervals of 30 degrees in the horizontal rotation angle θ and 30 degrees in the vertical direction with respect to the radius r, From each position, vertical, horizontal,
The rendering animation of the heart is stored for the movements in the right and left diagonal directions.

FIG. 27 is a plan view showing the arrangement of moving video data shot at rotation angles θ and δ of 30 degrees each. In FIG. 27, one cell is 30 in each of the vertical and horizontal directions.
Degrees. For example, if you move right from a point,
When moving to the left from the right point next to it, the same data can be used because the content of the moving video is the same and the playback is reversed. Only needs to be done.

In FIG. 27, for example, if the viewpoint is moved along a thick line, at intervals of 0.5 seconds,
Analyzing the movement of the searcher's viewpoint, selecting the moving video closest to the movement and displaying the moving video on the virtual panel. In the case of the example shown in FIG. 27, it indicates that the moving video between the grid points indicated by ● is selected.

Analysis of viewpoint movement and video selection will be described with reference to FIG. At some point, when the moving video is played from s to p, the next moving video is selected.
At this time, it is assumed that the viewpoint has moved from a to b in the previous time interval. Since the moving video ends with p, the next moving video may have eight directions starting from p. Therefore, the movement in these eight directions is compared with the movement from a to b. For example, as parameters, the inclination of the moving direction, the moving distance, and the proximity of the end point,
Those having a similar moving direction are selected, for example, three.
Among them, if the moving distance of a → b is less than half of the video movement, it is excluded from the selection. If there are no more options, the moving video is selected at the next timing. Next, the inclination of the direction is compared with the position of the end point, and a moving video in which both are closer is selected. In the case of FIG. 28, a → b → c
This indicates that the moving video of s → p → q → r has been selected for the viewpoint movement of.

As described above, a moving video closer to the searcher's viewpoint can be selected. However, when the next reproduced video turns at an angle of 90 degrees or 45 degrees, for example, FIG.
In the case of generating a video of p → q next to s → p in 8, the video becomes unnatural compared to the actual viewpoint movement.
Therefore, in the case where s → p → q is continuously played back, interpolation processing is performed using frames of the moving video around p, and the smooth moving video shown in FIG. 29 is played back. Thereby, a smooth viewpoint movement can be realized. FIG. 30 shows the addition of the moving video interpolation unit 78 for performing the interpolation processing.

As described above, even if CG objects and anchors corresponding to the respective items of the classification system are not prepared, the information of the upper and lower items is obtained while maintaining consistency with the hierarchical relationship of the classification system. By creating objects and anchors and defining link information, it is possible to support the work of the constructor and to construct a classification system in a three-dimensional space using CG objects, images, and moving videos, thereby reducing the work of the constructor. A book search interface compatible with the classification system can be constructed.

Further, a moving video of the composite object from the CG object is automatically generated, and the same expression as that of the CG object is possible in the three-dimensional search space. Can be searched. By selecting and displaying the moving video prepared for each viewpoint in accordance with the searcher's movement of the viewpoint, the searcher can freely walk around the object, similarly to the CG object.

Further, since the interval between grid points for capturing moving video can be easily changed, a smooth viewpoint movement according to the high definition required for the object is possible. When the interval between viewpoint movements is increased, the playback amount of the video at each time increases, and the viewpoint movement may be different from the viewpoint movement that moves finely, but the path of the viewpoint movement is roughly determined. This is particularly effective when it is sufficient to be able to roughly follow.

It should be noted that the combination of the above-described embodiments can be combined with the composite object according to the item based on the preparation status of the object of the item of the classification system, the judgment of the builder, or the relationship with the number of polygons of the object. It is also possible to build a classification system that mixes multiple media, such as generating a video frame object, generating a moving video, etc., and it is necessary to represent various classifications of books Can be effectively expressed.

In constructing video data, CG
The video frames and the moving video were shot for the object, but the same effect can be obtained not only by shooting from the CG object but also by using a video frame and a moving video that were actually shot. Therefore, it is possible to cope with a case in which a non-existent three-dimensional CG model is targeted for searching a book. For example, when trying to realize a human model with a CG object, a large number of polygons is required. However, by using the present invention, it is possible to use the data as a substitute for three-dimensional data with a small amount of data.

[0123]

As will be understood from the above description, according to the book information search space construction / search apparatus according to the present invention, a book information classification system is constructed and searched on a three-dimensional search space constructed on a computer. The searcher can select the CG object and image object-related anchors related to the book information that the user wants to search, follow the related information, and finally search for the necessary information. it can.

Construction of book information search space according to the following invention
According to the search device, even if a CG object corresponding to each item of the classification system or an anchor is not prepared, it is possible to obtain an object or an object from information of upper and lower items while maintaining consistency with the hierarchical relationship of the classification system. By generating anchors and defining link information, the work of the builder can be supported and a classification system in the three-dimensional search space can be constructed, so that the work of the builder can be reduced. A consistent book search interface can be constructed. Also,
By defining the hierarchical relationship between objects in the three-dimensional search space as concept keywords defined for each anchor instead of static link information, dynamic link information for searching for a link destination at the time of search can be realized. In the case where the content of the anchor or the anchor is updated, it is not necessary to update all the already linked information and the like, and the data can be managed independently.

Construction of book information retrieval space according to the following invention
According to the search device, a classification system of book information is constructed on a three-dimensional search space constructed on a computer and presented to a searcher, and the searcher can find a CG object, image object, By selecting an anchor by a video frame and tracing further related information, it is possible to search for finally necessary information.

Construction of book information retrieval space according to the following invention
According to the search device, a video frame is automatically generated from the CG object, and the same expression as the CG object can be performed in the three-dimensional search space. The object can be referred to by the video frame and related information can be searched. Therefore, even when the number of polygons of the CG object is large and a smooth walk-through is impossible, a smooth expression can be performed using the video frame as the substitute data. Also, the hierarchical relationship between objects in the three-dimensional search space is
By defining not a static link information but a concept keyword defined for each anchor, dynamic link information that searches for a link destination at the time of search can be realized, and when the contents of objects and anchors are updated, It is not necessary to update all linked information and the like, and data can be managed independently.

Construction of book information search space according to the following invention
According to the search device, video display can be performed according to the viewpoint, and a video frame can be photographed in accordance with the required fineness to realize a smooth expression.

Construction of book information retrieval space according to the following invention
According to the search device, it is possible to perform video display according to the viewpoint using frames obtained by the interpolation process, and realize a smooth expression with a small amount of photographing data.

Construction of book information search space according to the following invention
According to the search device, a classification system of book information is constructed on a three-dimensional search space constructed on a computer and presented to a searcher, and the searcher can find a CG object, image object, By selecting an anchor based on the moving video and tracing further related information, finally necessary information can be searched.

Construction of book information search space according to the following invention
According to the search device, the moving video is automatically generated from the CG object, and the same expression as the CG object can be performed in the three-dimensional search space. The object can be referred to by the moving video and related information can be searched. CG
Even when the number of polygons of the object is large and smooth walk-through is not possible, smooth representation can be performed using the moving video as substitute data. Also, by defining the hierarchical relationship between objects in the three-dimensional search space as concept keywords defined for each anchor instead of static link information, dynamic link information for searching for a link destination during a search can be realized. When the contents of an object or an anchor are updated, it is not necessary to update all the already linked information and the like, and the data can be managed independently.

Construction of book information search space according to the following invention
According to the search device, the moving video can be displayed according to the viewpoint, and the moving video can be captured according to the required fineness to realize a smooth expression.

Construction of book information search space according to the following invention
According to the search device, it is possible to display a moving video according to the viewpoint using frames obtained by the interpolation process,
Smooth expressions can be realized with less shooting data.

Construction of book information search space according to the following invention
According to the search device, related information can be searched using the attribute information of the selected anchor. In particular, since the concept keywords of the lower objects are set for a plurality of lower objects, a one-to-N Link definition (N> 1) and link execution become possible, and link execution like keyword search becomes possible.

Construction of book information search space according to the following invention
According to a computer-readable recording medium storing a search program, a classification system of book information is constructed on a three-dimensional search space constructed on a computer and presented to a searcher. It is possible to select an anchor by a related CG object or image object, follow further related information, and finally search for necessary information.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, even if the CG object corresponding to each item of the classification system and the anchor are not prepared, while maintaining consistency with the hierarchical relationship of the classification system, By defining objects and anchors and link information from lower-level item information, it is possible to support the work of the builder and build a classification system in the three-dimensional search space. A book search interface compatible with the classification system can be constructed. Also,
By defining the hierarchical relationship between objects in the three-dimensional search space as concept keywords defined for each anchor instead of static link information, dynamic link information for searching for a link destination at the time of search can be realized. In the case where the content of the anchor or the anchor is updated, it is not necessary to update all the already linked information and the like, and the data can be managed independently.

Construction of book information search space according to the following invention
According to a computer-readable recording medium storing a search program, a classification system of book information is constructed on a three-dimensional search space constructed on a computer and presented to a searcher. It is possible to select related CG objects, image objects, and anchors based on video frames, follow further related information, and finally search for necessary information.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, a video frame is automatically generated from the CG object, and the same expression as the CG object can be performed in the three-dimensional search space. In addition, since the related information can be searched, even when the number of polygons of the CG object is large and smooth walk-through is impossible, a smooth expression can be performed using the video frame as the substitute data. Also, the hierarchical relationship between objects in the three-dimensional search space is
By defining not a static link information but a concept keyword defined for each anchor, dynamic link information that searches for a link destination at the time of search can be realized, and when the contents of objects and anchors are updated, It is not necessary to update all linked information and the like, and data can be managed independently.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, video display can be performed according to the viewpoint, and video frames can be photographed according to the required fineness to realize a smooth expression.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, it is possible to perform video display according to the viewpoint using frames obtained by the interpolation processing, and realize a smooth expression with a small amount of photographing data.

Construction of book information search space according to the following invention
According to a computer-readable recording medium storing a search program, a classification system of book information is constructed on a three-dimensional search space constructed on a computer and presented to a searcher. It is possible to select related CG objects, image objects, and moving video anchors to trace further related information, and finally search for necessary information.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, the moving video is automatically generated from the CG object, and the same expression as the CG object can be expressed on the three-dimensional search space. And related information can be searched.
Even when the number of polygons of the object is large and smooth walk-through is not possible, smooth representation can be performed using the moving video as substitute data. Also, by defining the hierarchical relationship between objects in the three-dimensional search space as concept keywords defined for each anchor instead of static link information, dynamic link information for searching for a link destination during a search can be realized. When the contents of an object or an anchor are updated, it is not necessary to update all the already linked information and the like, and the data can be managed independently.

Construction of book information search space according to the following invention
According to the computer-readable recording medium storing the search program, the moving video can be displayed according to the viewpoint, and the moving video can be photographed according to the required fineness to realize a smooth expression.

Construction of a book information search space according to the following invention
According to the computer-readable recording medium storing the search program, it is possible to display the moving video according to the viewpoint using the frame obtained by the interpolation process,
Smooth expressions can be realized with less shooting data.

Construction of a book information search space according to the following invention
According to the computer-readable recording medium storing the search program, the related information can be searched using the attribute information of the selected anchor. In particular, the concept keyword of the lower object is set for a plurality of lower objects. As a result, one-to-N (N> 1) link definition and link execution can be performed, and link execution like keyword search can be performed.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing Embodiment 1 of a book information search space construction / search apparatus according to the present invention.

FIG. 2 is a block diagram illustrating details of an automatic CG search space construction unit of the book information search space construction / search apparatus according to the first embodiment;

FIG. 3 is a functional block diagram illustrating a processing outline of a correspondence setting unit of the book information search space construction / search apparatus according to the first embodiment;

FIG. 4 is a functional block diagram illustrating a processing outline of a CG object automatic construction unit of the book information search space construction / search apparatus according to the first embodiment;

FIG. 5 is a functional block diagram illustrating a processing outline of an automatic CG anchor construction unit of the book information search space construction / search device according to the first embodiment;

FIG. 6 is a functional block diagram illustrating a processing outline of an automatic attribute information setting unit of the book information search space construction / search apparatus according to the first embodiment;

FIG. 7 is an explanatory diagram showing an example of classification system information used in the book information search space construction / search apparatus according to the present invention.

FIG. 8 is an explanatory diagram showing an example of a classification item association interface in the book information search space construction / search apparatus according to the present invention.

FIG. 9 is an explanatory diagram showing an example of attribute information setting in the book information search space construction / search device according to the present invention.

FIG. 10 is a configuration diagram showing Embodiment 2 of a book information search space construction / search apparatus according to the present invention.

FIG. 11 is a block diagram illustrating details of an automatic CG / video search space construction unit of the book information search space construction / search apparatus according to the second embodiment.

FIG. 12 is a functional block diagram showing an outline of processing of a video frame automatic construction unit of the book information search space construction / search apparatus according to the second embodiment.

FIG. 13 is a functional block diagram illustrating an outline of processing of an automatic video frame anchor construction unit of the book information search space construction / search device according to the second embodiment.

FIG. 14 is an explanatory diagram showing an example of attribute information setting in the book information search space construction / search device according to the present invention.

FIG. 15 is a block diagram illustrating details of a video display processing unit of the book information search space construction / search apparatus according to the second embodiment.

FIG. 16 is a diagram illustrating a state in which a virtual panel is displayed according to a viewpoint.

FIG. 17 is an explanatory diagram showing a viewpoint of video shooting.

FIG. 18 is an explanatory diagram showing polar coordinates in video shooting.

FIG. 19 is an explanatory diagram showing shooting data of a video frame.

FIG. 20 is an explanatory diagram showing shooting data of a video frame.

FIG. 21 is a block diagram showing another example of the details of the video display processing means of the book information search space construction / search apparatus according to the second embodiment.

FIG. 22 is a block diagram illustrating details of a CG / video search space automatic construction unit of the book information search space construction / search apparatus according to the third embodiment.

FIG. 23 is a functional block diagram showing an outline of processing of a mobile video automatic construction unit of the book information search space construction / search apparatus according to the third embodiment.

FIG. 24 is a functional block diagram showing an outline of processing of a moving video anchor automatic construction unit of the book information search space construction / search apparatus according to the third embodiment.

FIG. 25 is an explanatory diagram showing an example of attribute information setting in the book information search space construction / search device according to the present invention.

FIG. 26 is a block diagram illustrating details of a video display processing unit of the book information search space construction / search apparatus according to the third embodiment.

FIG. 27 is an explanatory diagram showing a positional relationship between moving video and viewpoint movement.

FIG. 28 is an explanatory diagram showing a selection result of moving video with respect to viewpoint movement.

FIG. 29 is an explanatory diagram showing a case where a moving video is interpolated with respect to a viewpoint movement and displayed.

FIG. 30 is a block diagram showing another example of the details of the video display processing means of the book information search space construction / search apparatus according to the third embodiment.

FIG. 31 is a block diagram showing a conventional three-dimensional information search device.

[Explanation of symbols]

10, 30 search space construction unit, 11 data input means,
12 CG search space automatic construction means, 13 partial anchor construction means, 14, 14 'CG / video search space automatic construction means, 20 correspondence setting section, 21 CG object automatic construction section, 22 CG anchor automatic construction section, 23
Automatic attribute information setting unit, 24 automatic video frame construction unit, 25 automatic video frame anchor construction unit, 26
Moving video automatic construction unit, 27 moving video anchor automatic construction unit, 50 search unit, 51 display unit, 52 mouse pointing unit, 53 three-dimensional search unit, 54 attribute information search unit, 55 book search unit, 60 search unit, 56, 5
6 'video display processing means, 57 CG / video search means, 70 viewpoint determination section, 71 video frame selection section,
72 virtual panel / position setting section, 73, 77 video mapping section, 74 video frame interpolation section, 75 moving video selection section z, 76 virtual panel / position setting section, 78
Moving video interpolation unit, 100 classification system information database, 101 object database, 102 anchor / link attribute database, 103 book information database.

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Satoshi Tanaka 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsubishi Electric Corporation F-term (reference) 5B050 BA07 BA09 EA27 EA28 FA02 FA09 GA08 5B075 ND20 PP03 PP13 PQ02 5E501 AA01 AC23 AC33 BA03 BA05 CA02 CB09 DA06 EA05 FA03 FA14 FA27

Claims (22)

[Claims] 1. A search space for searching for desired book information while searching related information on a virtual three-dimensional search space built on a computer, and searching for book information. Is a book information search space construction / search apparatus that performs a book classification system necessary for building a three-dimensional search space for book information search and computer graphics objects and image object data in the three-dimensional space. Automatic search space construction for establishing a three-dimensional search space by associating a book classification system with a computer graphics object or image object in a three-dimensional space based on data input means to be inputted and data inputted by the data input means. Means for constructing a partial anchor for an object in a three-dimensional search space; A search space constructing unit, a display unit for displaying the three-dimensional search space constructed by the search space constructing unit, a pointing device for pointing the three-dimensional search space, and a selection by the pointing of the pointing device. Three-dimensional search means for searching for information on the link destination of the selected anchor, and searching for book information using book classification system information associated with the link destination of the anchor selected by the pointing instruction of the pointing device. And a search unit configured by a book search means.
Search device. 2. A correspondence setting unit for associating each item of the book classification system with an input object and determining a necessary construction and setting according to a correspondence situation. In the case where there is no object corresponding to the item of the book classification system, and if there are a plurality of objects in the lower item, an object of the upper item is automatically generated from the information thereof, and the object automatic constructing unit which associates the object. Anchor automatic construction unit that automatically constructs an anchor required as a link source for associating a higher-order item with a lower-order item in the book classification system, and an attribute of the anchor as information of the book classification system. 2. The book information search space construction / search according to claim 1, further comprising: an attribute information automatic setting unit that automatically sets the information as information. Apparatus. 3. A search space for searching for target book information while searching for related information on a virtual three-dimensional search space built on a computer, and searching for book information. Is a book information search space construction / search apparatus that performs a book classification system necessary for building a three-dimensional search space for book information search and computer graphics objects and image object data in the three-dimensional space. Data input means for inputting, a classification system of books and a computer graphics object on a three-dimensional space according to data input by the data input means;
Automatic search space construction means for constructing a three-dimensional search space by associating image objects or video frame objects with each other, and partial anchor construction means for constructing a partial anchor for objects in the three-dimensional search space Unit, display means for displaying the three-dimensional search space constructed by the search space construction part, and video display processing means for performing processing for displaying a video frame on a virtual panel, which is a single polygon in the three-dimensional search space. A pointing device that points to the three-dimensional search space; a three-dimensional search unit that searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device; and an anchor that is selected by the pointing instruction of the pointing device. Books linked to links in And a search unit comprising book search means for searching for book information using the classification system information of
Search device. 4. An automatic search space construction apparatus, comprising: a correspondence setting unit for associating each item of a book classification system with an input object and determining a necessary construction and setting according to a state of association; Using the computer graphics object of the lower item or the computer graphics object of its own, the computer graphics object is photographed at multiple viewpoints, and a video frame that is a set of the rendering images for each viewpoint is automatically generated. A video frame automatic construction unit for generating and registering in a database so that viewpoint movement in multiple directions is possible in a three-dimensional search space instead of a computer graphics object, and associating the object with an item; To lower level items To do
It includes a video frame / anchor automatic construction unit that automatically constructs an anchor required as a link source for a video frame object, and an attribute information automatic setting device that automatically sets information as attribute information of the anchor. 4. The book information search space construction / search apparatus according to claim 3, wherein: 5. A video display processing unit comprising: a viewpoint determination unit for determining a viewpoint of a searcher; a video frame selection unit for selecting a video frame from a database in accordance with the position of the viewpoint; A virtual panel / position setting unit that determines a position of a virtual panel, which is a single polygon in a three-dimensional search space for mapping a frame, and a video mapping unit that maps a selected video frame to the virtual panel. The book information search space construction / search apparatus according to claim 3 or 4, wherein the book information search space construction / search apparatus is included. 6. The video display processing means further obtains, at a viewpoint position at which a video frame is not prepared, interpolation processing from data of surrounding video frames already prepared and uses the data as a selected frame. The book information search space construction / search apparatus according to claim 5, further comprising a frame interpolation unit. 7. A search space for searching for desired book information while searching for related information in a virtual three-dimensional search space built on a computer, and searching for book information. Is a book information search space construction / search apparatus that performs a book classification system necessary for building a three-dimensional search space for book information search and computer graphics objects and image object data in the three-dimensional space. Data input means for inputting, a classification system of books and a computer graphics object on a three-dimensional space according to data input by the data input means;
Search space construction means comprising automatic search space construction means for constructing a three-dimensional search space by associating image objects or moving video objects with each other, and partial anchor construction means for constructing partial anchors for objects in the three-dimensional search space. Unit, display means for displaying the three-dimensional search space constructed by the search space construction part, and video display processing means for performing processing for displaying a moving video on a virtual panel, which is a single polygon in the three-dimensional search space. A pointing device that points to the three-dimensional search space; a three-dimensional search unit that searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device; and an anchor that is selected by the pointing instruction of the pointing device. Classification of books associated with the link destination of And a search unit comprising a book search means for searching for book information using the system information.
Search device. 8. The automatic search space construction device, comprising: a correspondence relationship setting unit that associates each item of the book classification system with an input object and determines a necessary construction and setting according to a correspondence situation; Using the lower-level computer graphics object or the computer graphics object of its own, a moving video that captures the viewpoint movement of the computer graphics object from a plurality of viewpoints is automatically generated to generate the computer graphics object. Instead, a moving video automatic construction unit that registers the object in the database so that the viewpoint can be moved in a plurality of directions on the three-dimensional search space and associates the object with the item, and associates the upper item with the lower item. To do
A moving video / anchor automatic construction unit for automatically constructing an anchor required as a link source for a moving video object; and an attribute information automatic setting device for automatically setting information as attribute information of the anchor. 8. The book information search space construction / search apparatus according to claim 7, wherein: 9. The video display processing means, a viewpoint determining unit for determining a viewpoint of a searcher, a moving video selecting unit for selecting a moving video from a database according to the position of the viewpoint, and moving according to the position of the viewpoint. A virtual panel / position setting unit for determining a position of a virtual panel, which is a single polygon in a three-dimensional search space for mapping a video, and a video mapping unit for mapping the selected moving video to the virtual panel. 9. The book information search space construction / search apparatus according to claim 7, wherein the book information search space is constructed. 10. The video display processing means further comprises:
At a viewpoint position where a moving video is not prepared, a moving video interpolating unit is obtained by interpolating from data of surrounding moving videos already prepared and using the data as a selected moving video. Item 9. A book information search space construction / search device according to item 9. 11. The apparatus according to claim 1, wherein the search unit includes an attribute information search unit that searches related information from attribute information of the anchor selected by the pointing instruction of the pointing device. The book information search space construction / search device according to any one of the above. 12. A search space for searching for desired book information while searching for related information on a virtual three-dimensional search space built on a computer, and searching for book information. A computer-readable recording medium having recorded thereon a program for performing a three-dimensional correspondence between a book classification system and a computer graphics object or image object in a three-dimensional space by data input by data input means. A search space construction procedure for constructing a search space and constructing a partial anchor for an object in the three-dimensional search space, and a three-dimensional search space constructed by executing the search space construction procedure are displayed, and selected by a pointing instruction of a pointing device. Searches for the linked information of the anchor that was Computer readable recording medium storing a search procedure, the which utilizes a classification scheme information of books associated with the anchor of the link destination selected by preparative instruction to search for books information. 13. The search space construction procedure associates each item of the book classification system with an input object, determines necessary construction and settings according to the association status, and determines the book classification system. If there is no object corresponding to the item and there are multiple objects in the lower item,
Based on the information, objects of higher items are automatically generated and associated, and anchors required as link sources for associating upper items with lower items in the book classification system are automatically generated. 13. The computer-readable recording medium storing a program according to claim 12, wherein the information in the book classification system is automatically set as attribute information of the anchor. 14. A search space for searching for desired book information while searching for related information on a virtual three-dimensional search space built on a computer, and searching for book information. A computer-readable recording medium on which a program to be recorded is recorded, wherein a book classification system is associated with a computer graphics object, image object, or video frame object in a three-dimensional space by data input by data input means. A search space construction procedure for constructing a three-dimensional search space, and constructing a partial anchor for an object in the three-dimensional search space; and displaying the three-dimensional search space constructed by executing the search space construction procedure. Performs processing to display on a virtual panel that is a single polygon in the dimensional search space, Searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device, and uses the book classification system information associated with the link destination of the anchor selected by the pointing instruction of the pointing device to retrieve the book information. A computer-readable recording medium on which is stored a search procedure for searching. 15. The search space construction procedure associates each item of a book classification system with an input object, determines necessary construction and settings according to the state of association, and determines a computer of a lower item. Using a graphics object or your own computer graphics object to take a rendered image of the computer graphics object from multiple viewpoints,
A video frame, which is a set of rendered images for each viewpoint, is automatically generated and registered in a database so that viewpoints can be moved in a plurality of directions on a three-dimensional search space in place of a computer graphics object, and the object is registered. Automatically constructs an anchor required as a link source for video frame objects in order to associate items with higher items and lower items, and automatically converts information as attribute information of anchors. A computer-readable recording medium on which the program according to claim 14 is set. 16. The three-dimensional search space for determining a viewpoint of a searcher, selecting a video frame from a database according to the position of the viewpoint, and mapping the video frame according to the position of the viewpoint. 16. The computer-readable recording medium according to claim 14, wherein a position of the virtual panel, which is one polygon, is determined, and a selected video frame is mapped to the virtual panel. 17. The search procedure according to claim 7, further comprising, at a viewpoint position where a video frame is not prepared, obtaining data by interpolation from data of surrounding video frames already prepared and using the data as a selected frame. A computer-readable recording medium on which the program according to claim 16 is recorded. 18. A search space for searching for desired book information while searching for related information in a virtual three-dimensional search space built on a computer, and searching for book information. A computer-readable recording medium on which a program to be recorded is recorded, wherein a book classification system is associated with a computer graphics object, image object, or moving video object in a three-dimensional space by data input by data input means. A search space construction procedure for constructing a three-dimensional search space, and constructing a partial anchor for an object in the three-dimensional search space; and displaying the three-dimensional search space constructed by executing the search space construction procedure. Performs processing to display on a virtual panel, which is a single polygon in the three-dimensional search space, and Searches for information on the link destination of the anchor selected by the pointing instruction of the pointing device, and searches for book information using the classification system information of the book associated with the link destination of the anchor selected by the pointing instruction of the pointing device A computer-readable recording medium that records the search procedure and 19. The search space construction procedure associates each item of the book classification system with an input object, determines necessary construction and settings according to the state of association, and determines a computer of a lower item. Using a graphics object or its own computer graphics object, automatically generate a moving video of the viewpoint movement of the computer graphics object from a plurality of viewpoints, and a three-dimensional search space instead of the computer graphics object Register in the database so that the viewpoint can be moved in multiple directions above, move the anchor required as a link source for associating the object with the item and associating it from the upper item to the lower item Automatically builds and informs about video objects The computer-readable recording medium according to claim 18, wherein the information is automatically set as attribute information of the anchor. 20. The search procedure, comprising: determining a viewpoint of a searcher; selecting a moving video from a database according to the position of the viewpoint; and mapping the moving video according to the position of the viewpoint on a three-dimensional search space. 19. The position of the virtual panel, which is a single polygon, is determined, and the selected moving video is mapped to the virtual panel.
Or a computer-readable recording medium recording the program according to 19. 21. The search procedure further includes, at a viewpoint position where a moving video is not prepared, obtaining data by interpolation from data of surrounding moving videos already prepared and using the data as a selected moving video. A computer-readable recording medium on which the program according to claim 20 is recorded. 22. The program according to claim 12, wherein the search step searches for related information from attribute information of an anchor selected by a pointing instruction of a pointing device. A computer-readable recording medium that has been recorded.