JP2007072918A - Information retrieval system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information retrieval system capable of extracting an important image from accumulated pickup moving images. <P>SOLUTION: The information retrieval system is equipped with: an analysis device 20 for analyzing an electronic slide and extracting information regarding an object included in the slide; an imaging device 30 for imaging the slide so that an image can be captured; an indicated position detecting device 41 for detecting the indicated position of the displayed slide; an object specifying means 210 for specifying an indicated object from information concerning the indicated position and the object; a time-series database 320 for associating and accumulating the indicated object with the pickup image; and a retrieving means 430 for retrieving the desired image from the pickup image accumulated in the time-series database 320. By this structure, an image can be retrieved based on the object indicated by a speaker. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information retrieval system capable of retrieving an important image from accumulated captured images.

By replaying the video recording of the lecture, the state of the lecture can be seen and further reconfirmed. At this time, if the scene that the lecturer thinks is important can be extracted in advance before playback, the contents of the video recording of the lecture can be confirmed efficiently, and the abstract of the lecture can be grasped in a short time. can do. As an information retrieval system having such a function, a technique for accumulating instructed coordinates of the pointing device in association with moving image information or material information and retrieving a scene that the lecturer considers important is known (for example, Patent Document 1).
JP 2003-330935 A

By the way, in the information retrieval system as described above, an important scene can be specified in units of slides, but it is specifically estimated which character string, figure, table or other object in the electronic slide is important. I couldn't.
For this reason, a scene of a captured image (hereinafter referred to as a “scene”) cannot be accurately searched based on an important object.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an information search system that can identify important objects and allow a speaker to search for scenes that explain important objects. It is to provide.

An information search system according to the present invention is applied to an analysis unit that analyzes a slide displayed based on electronic material and extracts an object included in the slide, and an object extracted by the analysis unit and a displayed slide. The object specifying means for specifying the pointing object specified by the pointing operation by comparing the pointing position information based on the pointing operation, the captured image including the displayed slide, and the pointing object specified by the object specifying means. It is characterized by comprising time series storage means for storing in association with each other and search means for searching for a desired scene from captured images stored in the time series storage means based on the pointing object.
According to this configuration, a scene can be searched based on the instruction object.

In the above configuration, the analysis means determines information related to the object range that is a range in which the speaker determines that the object is specified, and the object specifying means specifies the specified object from the information related to the object range determined by the analysis means. Can be adopted.
According to this configuration, it becomes easy for the information system to determine whether the speaker has designated the object.

The said structure WHEREIN: The structure further provided with the result display means to display the scene searched by the search means is employable.
According to this configuration, the scene searched by the operator can be visually recognized.

In the above configuration, as a result, it is possible to adopt a configuration in which thumbnails of searched scenes are displayed.
According to this configuration, an overview of the searched scene can be known at a glance.

In the above-described configuration, there is further provided an importance level specifying unit that determines the importance level of the slide based on the indication object stored in the time series storage unit and the time information that is the specified time of the indication object.
A configuration further including importance display means for visually displaying the importance of the slide determined by the importance determination means can be adopted.
According to this configuration, the operator can know the importance of the slide at a glance.

In the above configuration, the importance level display means may be configured to display a slide thumbnail.
According to this configuration, the operator can know the outline of the slide at a glance.

In the above configuration, the image processing apparatus further includes an importance level specifying unit that determines the importance level of the object based on the indication object and time information stored in the time series storage unit, and the importance level of the object determined by the importance level determination unit is visually It is possible to employ a configuration that further includes importance display means for displaying automatically.
According to this configuration, the operator can know the importance of the object at a glance.

In the above configuration, it is possible to adopt a configuration further including condition setting means for setting the search condition of the search means.
According to this configuration, the operator can search for a scene based on various search conditions.

In the above configuration, the analysis unit extracts information on the type of the object, the object specification unit specifies the type of the instruction object from the information on the type of object extracted by the analysis unit, and the search unit determines the type of the instruction object. A configuration for searching for a scene based on the above can be adopted.
According to this configuration, for example, the operator can search only the scene in which the speaker indicates the graphic object.

In the above configuration, the analysis unit can add or delete the extracted object, and can add, delete, or change information about the object range or information about the type of the object.
According to this configuration, it is difficult for a difference between the judgment of the present information system that the speaker has instructed the object and the judgment of the speaker himself.

In the above configuration, the importance level determination unit may employ a configuration in which a slide or an object uses a deviation value of a cumulative value of time instructed by a speaker as an index indicating the level of importance.
According to this configuration, since the importance level is determined based on the behavior of the speaker, a difference between the importance level considered by the speaker and the importance level used by the present system hardly occurs.

In the above configuration, the importance level determination unit may employ a configuration in which the deviation value of the scoring result value corresponding to the positional relationship between the designated position and the displayed object is used as an index indicating the importance level.
According to this configuration, even for a material having a more complicated object configuration, a difference between the importance level determined by the speaker and the importance level used by the present system hardly occurs.

In the above configuration, the search means can employ a configuration for searching for an important scene based on a duration time in which the lecturer indicates the object.
According to this configuration, for example, the operator can search for a scene in which a speaker has instructed a specific object for 5 seconds or longer.

In the above configuration, the search means can employ a configuration for searching for a scene based on the accumulated time when the lecturer has indicated the object.
According to this configuration, for example, the operator can search for a series of scenes in which a lecturer points to a specific object, and the total time length of each scene exceeds 10 seconds.

In the above configuration, it is possible to further include a speaker management unit that stores information for identifying the speaker in the time series storage unit, and the search unit can employ a configuration for searching for a scene based on the information for identifying the speaker.
According to this configuration, a scene can be searched based on a speaker.

The above configuration further includes position detection means for detecting the position of the pointing device used for the participant and the speaker to indicate the displayed slide, and the speaker management means is based on the positions of the participant and the pointing device. , You can adopt a configuration that guesses the speaker from the participants.
According to this configuration, since this information system automatically accumulates who the speaker is, the operation burden on the operator can be reduced.

  According to the present invention, a stored image captured in the past can be searched based on an object, so that a lecturer can search for a scene explaining a specific object.

DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, exemplary embodiments of the invention will be described with reference to the accompanying drawings.
FIG. 1 is a diagram showing a configuration of an information system according to an embodiment of the present invention and data exchange between devices.
As shown in FIG. 1, the information system includes a data collection unit 100, a data totaling unit 200, a data storage unit 300, and a data display unit 400.

The data collection unit 100 includes a display device 10, an analysis device 20 as analysis means, a video camera 30, an instruction device 40, a position detection device 50 as position detection means, and an information management device 60 as lecturer management means.
The data collection unit 100 is connected to the data storage unit 300.

The display device 10 includes a screen 11, a projector 12, and a slide control device 13.
The screen 11 is composed of, for example, a bead screen. The screen 11 projects a slide by the projector 12.

The projector 12 is composed of, for example, a liquid crystal projector. The projector 12 is connected to the slide control device 13.
The projector 12 projects the slide transmitted from the slide control device 13 onto the screen 11.

Here, before describing the slide control device 13, the relationship among the material, the slide, the object, the object range, and the object type will be described.
FIG. 2 is a diagram for explaining the relationship between a material, a slide, an object, and an object range.

As shown in FIG. 2, the material D is electronic data composed of one or a plurality of slides S. The document D is created by, for example, presentation software called Microsoft PowerPoint.
The slide S includes one or more objects OB or no objects OB at all.
The object OB is a character string, figure, or table that constitutes the slide S.
In addition, any of character strings, diagrams, and tables that are object types is referred to as an object type.
Further, a range that is a predetermined range and is used to determine that the object OB is instructed when the speaker indicates the range is referred to as an object range ROB.
In the following, it is assumed that the object ranges ROB do not overlap unless otherwise specified.

The slide control device 13 is composed of, for example, a personal computer. The slide control device 13 is connected to the projector 12, the analysis device 20, and the information management device 60.
First, an operator inputs electronic materials into the slide control device 13. Thereafter, the slide control device 13 stores the input electronic material and determines information for identifying the material (hereinafter referred to as material ID) and information for identifying the slide (hereinafter referred to as slide ID).

Next, the input electronic material, material ID, and slide ID are transmitted to the analysis device 20.
Further, the stored slide is transmitted to the projector 12 according to the operation of the operator or automatically, and the slide ID of the electronic slide transmitted to the projector 12 and the material ID of the material including the slide are transmitted to the information management device 60. To do.

  The analysis device 20 is configured by, for example, a personal computer. The analysis device 20 is connected to the slide control device 13 and the material database 310 of the data storage unit 300.

  The analysis device 20 analyzes the electronic material received from the slide control device 13, determines information for identifying the object (hereinafter referred to as object ID), information on the object range (hereinafter referred to as object range information), and the object Extract the type.

The manner in which the analysis device 20 analyzes the material will be described with reference to FIG.
FIG. 3 is a diagram for explaining information that the analysis apparatus has before and after the slide analysis.
As shown in FIG. 3, the analysis device 20 has information called the material D and the slide S, the material ID IDD, and the slide ID IDS received from the slide control device 13 before analysis.

  Thereafter, the description content of the file is analyzed based on the file format of the material D, the object OB and the object type KOB included in each slide S are extracted, and the object ID IDO and the object range ROB are determined.

Here, how the analysis apparatus 20 extracts the object range information will be described with reference to FIG.
FIG. 4 is a diagram for explaining a method of extracting object range information.
As shown in FIG. 4, it is assumed that the object range ROB is square in this embodiment. Further, the object range ROB is displayed using two two-dimensional coordinates of the vertex VCN closest to the origin O and the vertex CVF farthest from the origin.

Thus, in order for the information system to determine whether the lecturer has designated the object, the position coordinates on the display slide (hereinafter referred to as designated position information) where the lecturer has designated the vertex VCN and the vertex CVF are displayed on the electronic slide. What is necessary is just to compare with what was converted into the position coordinates (hereinafter referred to as the slide instruction position).
This makes it easy for the information system to determine whether the speaker has indicated the object.

  Further, when necessary, the analysis apparatus 20 adds, deletes, or corrects object ID, object range information, or information on an object type (hereinafter referred to as object type information) by an operator's operation or automatically.

As a result, for example, when the object range that did not contain the object is enlarged and corrected so as to include the object range, the speaker does not directly indicate the object, The system can recognize that an object has been indicated.
Therefore, the difference between the judgment of the information system that the lecturer has instructed the object and the judgment of the lecturer is less likely to occur.

The manner in which the analysis device 20 corrects the object identification information, the object range information, or the object type information will be described with reference to FIGS.
FIG. 5 is a diagram for explaining a method for dividing and combining objects, and FIG. 6 is a diagram illustrating a method for correcting information related to objects.

As illustrated in FIG. 5, the analysis apparatus 20 extracts the object OB8.
A case where the extracted one object OB8 is divided into two upper and lower objects OB8 and OB9 will be described.

First, the operator changes information related to the object range ROB8 of the object OB8 by using the analysis device 20.
Thereafter, information related to the object ID IDO9, the object range ROB9, and the object type KOB9 relating to the object OB9 is newly added.
Thereby, the object OB8 can be divided.

Conversely, a case where the extracted object OB8 and object OB9 are combined to form one object OB8 will be described.
The operator changes information related to the object range ROB8 of the object OB8 using the analysis device 20.
Thereafter, the information regarding the object ID IDO9, the object range ROB9, and the object type KOB9 regarding the object OB9 is deleted.
Thereby, the object OB8 and the object OB9 can be combined.

Here, as an example of a method for correcting information related to an object, as shown in FIG. 6, there is a method of dividing an slide into a plurality of regions and adding an object having each region range as an object range.
By correcting in this way, it is possible to extract the importance for each slide area and search for an important scene based on the slide area.

  Further, the analysis device 20 stores the material, the material ID and the slide ID received from the slide control device 13, the object ID, the object range information, and the object type information as the analysis result in the material database 310.

Here, the manner in which the analysis apparatus 20 stores various types of information in the material database 310 will be described with reference to FIG.
FIG. 7 is a configuration diagram of an analysis result table included in the material database.
As shown in FIG. 7, the analysis apparatus 20 accumulates the analysis result information of the electronic material in the analysis result table TA of the material database 310.

  The analysis result table TA stores a material column CD that stores material IDs, a slide column CS that stores slide IDs, an object column CO that stores object IDs, an object type column CKO that stores object type information, and object range information. The object range column CRO includes a cumulative time column CAT for accumulating the time for which the lecturer has designated the object and totaled for each object, and a deviation value column CAD for accumulating deviation values of the cumulative time.

The analysis apparatus 20 accumulates the information in the material column CD, the slide column CS, the object column CO, the object type column CKO, and the object range column CRO. Further, the importance level identifying device 220 of the data totaling unit 200 stores the remaining accumulated time column CAT and deviation value column CAD.
Note that the analysis result table TA does not have duplicate records in which the data in the material column CD and the object column CO all match.

The video camera 30 is composed of, for example, a CCD camera. This video camera 30 is connected to the information management device 60.
The video camera 30 is arranged so as to be capable of capturing at least an image projected by the projector 12 on the screen 11.
The video camera 30 transmits the captured image to the information management device 60.

The instruction device 40 is configured by, for example, a laser pointer. The instruction device 40 is connected to the information management device 60.
The indicated position detection device 41 includes a position detection device 41 that is a position detection unit.

The designated position detection device 41 detects designated position information. As the detection technique, for example, the technique disclosed in Japanese Patent Application Laid-Open No. 2004-310527 is used.
In addition, the designated position information is transmitted to the information management device 60.

The pointing position detection device 41 has a pair of orthogonal acceleration sensors.
This acceleration sensor detects the magnitude of acceleration applied to the designated position detection device 41 when the designated position detection device 41 is moved.

Here, how the designated position detection device 41 detects the designated position will be described with reference to FIG.
FIG. 8 is a diagram for explaining a position detection method of the indicated position detection device.
First, the pointing position detection device 41 inserts the corner CN1 of the screen 11.
Thereafter, the designated position detection device 41 is operated to move the designated position from the corner CN1 toward the corner CN2. At this time, the indicated position detection device 41 is operated so that the indicated position moves on a straight line connecting two corners.

  Thereby, the indicated position detection device 41 stores the distance and the positional relationship between CN1 and CN2. Similarly, distances and positional relationships are stored for CN1 and CN3.

  Thereafter, the pointing position detection device 41 detects the pointing position of the speaker based on the magnitude and direction of the acceleration detected by the sensor, the acceleration time, and the positional relationship of the four corners of the screen 11 stored by the pointing position detection device 41. To do.

The position detection device 50 is connected to the information management device 60.
The position detection device 50 detects information (hereinafter referred to as a participant ID) for identifying a lecture participant, the position of the participant, and the position of the pointing device.
In addition, the participant ID, the detected participant position, and the pointing device position are transmitted to the information management device 60.

Here, how the position detecting device 50 detects the participant position and the pointing device position will be described with reference to FIG.
FIG. 9 is a diagram illustrating an arrangement of participants and instruction devices in a lecture hall or the like.
As shown in FIG. 9, the participant has an RFID with a unique ID and participates in the lecture, and also gives the RFID to the pointing device. By installing a plurality of readers at the venue, the position of each RFID and the unique ID of the RFID are detected.
Thereby, the position of the participant and the position of the pointing device are detected.

  The information management device 60 is constituted by a personal computer, for example. The information management device 60 is connected to the slide control device 13, the video camera 30, the pointing device 40, the position detection device 50, and the time series database 320 of the data storage unit 300.

In the information management device 60, information for identifying a lecture (hereinafter referred to as a lecture ID) is input by an operator. This is due to consideration of giving multiple lectures on the same material.
In addition, the information management device 60 converts the indication position received from the indication device 40 into a slide indication position.

Further, the participant ID, the participant position, and the pointing device position transmitted from the position detecting means 50 are received, and the speaker is identified from the participants.
Also, information for identifying a speaker (hereinafter referred to as a speaker ID) is determined.
As a result, the information system automatically accumulates who the speaker is, so the operation burden on the operator can be reduced.

Specifically, as illustrated in FIG. 9, the information management device 60 obtains a distance L between the RFID attached to each participant P and the RFID attached to the instruction device. These are obtained from the position coordinates of each RFID.
Thereafter, the lengths of the distances L are compared, and the participant P1 with the shortest distance L is estimated as a speaker.

Thereafter, the lecture ID and the lecturer ID, the material ID and the slide ID received from the slide control device 13, the captured image received from the video camera 30, and the slide indication position and time information received from the indication position detection device 70 are displayed. Accumulate in the time series database 320.
In the present embodiment, the time series database 320 is accumulated every second.

Here, how the information processing apparatus 60 stores various types of information in the time-series database 320 will be described with reference to FIG.
FIG. 10 is a configuration diagram of a time series table included in the time series database.
As shown in FIG. 10, the information management device 60 accumulates various types of information in the time series table TT of the time series database 320.

  As shown in FIG. 10, the time series table TT includes a time column CT for storing time information, a lecture column CL for storing lecture ID, a material column CD for storing material ID, and a speaker column CP for storing speaker ID. , A captured image column CM that stores information for identifying the stored captured image, a slide column CS that stores a slide ID, an instruction position column CPP that stores a slide instruction position, an object designated by a speaker (hereinafter referred to as an instruction object) ), An object column CO for storing an object ID (hereinafter referred to as an instruction object ID) and an object type column CKO for storing information on the type of the instruction object (hereinafter referred to as an instruction object type).

The information management device 60 stores information in the time column CT, the lecture column CL, the material column CD, the lecturer column CP, the captured image column CM, the slide column CS, and the indication position column CPP. Further, the object specifying device 210 of the data totaling unit 200 stores the remaining object column CO and object type column CKO.
Note that the time series table TT does not have duplicate records with the same data in the time column CT.

The data totaling unit 200 includes an object specifying device 210 as an object specifying means and an importance specifying device 220 as an importance specifying means.
The data totaling unit 200 is connected to the data storage unit 300.
The object specifying device 210 is connected to the material database 310 and the time series database 320.

The object specifying device 210 specifies an indicated object or the like from the slide instruction position stored in the time series database 320 and the object range information stored in the material database 310.
Thereafter, information about the pointing object is accumulated in the time series database 320.

Here, how the object specifying device 210 specifies the pointing object or the like from the slide pointing position or the like will be described with reference to FIG.
FIG. 11 is a diagram showing the relationship between the time-series database and the material database and the object specifying device.

As illustrated in FIG. 11, the object identifying device 210 refers to the information accumulated in the time series table TT included in the time series database 320 and the analysis result table TA included in the material database 310, and identifies an instruction object or the like.
Thereafter, the specified pointing object or the like is accumulated in the time series table TT.

Specifically, the object specifying device 210 searches for a record in the analysis result table TA that satisfies a certain condition with respect to the record in the time series table TT.
The certain condition is that the material IDs stored in the respective material columns CD of the time series table TT and the analysis result table TA match, and the slide IDs stored in the respective slide columns CS match, and The condition is that the object range information accumulated in the object range column CRO of the analysis result table TA includes the slide designation position accumulated in the designation position column CPP of the time series table TT.
Here, in this embodiment, since the object ranges do not overlap each other, there is no record that satisfies the above condition, or there is only one record.

  If there is a record that satisfies the condition, it can be seen that the speaker has instructed the object identified by the object ID accumulated in the retrieved record, and if there is no record, no object has been specified. I understand.

Thereafter, the object specifying device 210 refers to the object ID and the object type from the record of the analysis result table TA that satisfies the above conditions.
Further, the reference information is stored in the object column CO and the object type column CKO of the time series table TT.

  However, if no record satisfying the above condition exists in the analysis result table TA, data “000” is stored in the object column CO, and data “−” is stored in the object type column CKO ( Hereinafter, the object ID includes “000” and the object type includes “−”).

In order to explain more specifically, the first record of the time series table TT will be described as an example.
The record has data of a material ID “00001”, a slide ID “001”, and a slide instruction position “0, 0”.
The object specifying device 210 searches the analysis result table TA for records having the object range information including the material ID “00001”, the slide ID “001”, and the slide designation position “0, 0”. .
As a result of the search, it can be seen that such a record does not exist.

  Thereafter, the object identification device 210 accumulates data “000” in the object column CO of the first record of the time series table TT and data “−” in the object type column CKO.

Next, the second record of the time series table TT will be described as an example.
The record has data of a material ID “00001”, a slide ID “001”, and a slide instruction position “15, 18”.
The object specifying device 210 searches the analysis result table TA for records having the object range information including the material ID “00001”, the slide ID “001”, and including the slide instruction position “15, 18”. .

As a result of the search, it can be seen that the first record of the analysis result table TA satisfies the above condition.
This is because the first record of the analysis result table TA has the material ID “00001”, the slide ID “001”, and the object range information “15,15-175,35”.

Thereafter, the object identification device 210 refers to the object column CO and the object type column CKO stored in the record record that satisfies the search condition.
Further, the referenced object ID “001” and object type “C” are stored in the object column CO and the object type column CKO of the second record of the time series table TT.
These can be realized by describing a query or the like using a language such as SQL.

Here, a case where the object ranges overlap will be described.
When the object ranges overlap, two or more records may satisfy the search condition.
In that case, the object IDs stored in the respective records are stored in the object column CO of the time series table TT in a comma delimited manner.
Thereby, it is possible to cope with the case where the object ranges overlap. The same applies to the object type.

The importance specifying device 220 is connected to the material database 310 and the time series database 320.
The importance level identifying device 220 obtains a cumulative value of the time designated by the speaker for each slide and for each object as an index representing the importance level of the slide or object, and uses the deviation value. This is because the lecturer gives explanations while pointing to the object that is the explanation part, and takes time to explain the content that is considered important.

Here, how the importance level identifying device 220 counts the accumulated time and the deviation value will be described with reference to FIGS.
FIG. 12 is a diagram for explaining the importance specifying device, FIG. 13 is a diagram for explaining the importance specifying method for each slide, and FIG. 14 is a diagram for explaining the importance specifying method for each object. FIG.

  As shown in FIG. 12, the importance level specifying device 220 aggregates the records of the time series table TT of the time series database 320, and calculates the accumulated time for each slide and its deviation value, and the accumulated time for each object and its deviation value. Ask. Thereafter, the accumulated time and deviation value are accumulated in the slide table TS and the analysis result table TA of the material database 310.

First, a method in which the importance level identifying device 220 counts the accumulated time for each slide will be specifically described with reference to FIG.
A record in which the material ID stored in the material column CD and the slide ID stored in the slide column CS all match and the instruction object ID is not “000” is searched.
Thereafter, the accumulated time is calculated by multiplying the number of retrieved records by the data sampling interval time. The sampling interval time is the time interval for data recording, and here it is 1 second.

Next, the importance specifying device 220 stores the accumulated time for each slide in the accumulated time column CAT in the slide table TS of the material database 310 in association with the material ID and the slide ID.
Note that the slide table TS does not have duplicate records in which all the data in the material column CD and the slide column CS match.

  Thereafter, the importance level identifying apparatus 220 calculates the deviation value of the cumulative time after totaling the cumulative time for all the slides. Thereafter, the deviation value is accumulated in the deviation value column CAD of the slide table TS in association with the material ID and the slide ID.

Next, a method in which the importance level identifying device 220 counts the accumulated time for each object will be specifically described with reference to FIG.
As shown in FIG. 14, the importance level identifying device 220 has all the material IDs accumulated in the material column CD and the instruction object ID accumulated in the instruction object column CO from the records of the time series table TT of the time series database 320. A record that matches and that does not have the specified object ID “000” is searched.
Thereafter, the accumulated time can be calculated by multiplying the number of retrieved records by the data accumulation interval time.

Next, the importance specifying device 220 accumulates the accumulated time for each object in the accumulated time column CAT of the analysis result table TA.
Thereafter, after obtaining the accumulated time for all the objects, the deviation value of the accumulated time is calculated. Further, the deviation value is accumulated in the deviation value column CAD of the analysis result table TA.
These can be realized by describing a query or the like using a language such as SQL.

As a result, since the importance is determined based on the act of the speaker, it is less likely to differ from the importance considered by the speaker than the method of determining the importance from the number of nods of the participants and the gaze time.
Furthermore, since only the pointing position detection device and the RFID are used by the lecturer and the participant, not only a large-scale preparation is required, but also the burden on the lecturer is small.
Further, unlike the method of determining the importance level from the number of times the speaker has instructed the slide or the object, the speaker is less affected by noise or the like when the speaker plays with hands.

Here, a method of determining the importance when the object ranges overlap each other will be described with reference to FIG.
FIG. 15 is a diagram illustrating a method of determining the importance when the object ranges overlap each other.
As shown in FIG. 15, the object range ROB1 of the object OB1 and the object range OB3 of the object OB3 overlap each other.
Here, the slide instruction position PP1 is located only within the object range ROB1 at the time information T1, is located within both the object range ROB1 and the object range ROB2 at the time T2, and is located only within the object range ROB2 at the time T3. .

In such a case, the importance level specifying device 220 scores according to the positional relationship between the slide instruction position PP and the object OB.
Further, the score is stored in the score table TSC of the time series database 320 in association with the time information T.

Time information T is stored in the time column CT of the score table TSC, and the score of the object OB is stored in the score column CSC.
Thereafter, the importance level identifying device 220 calculates a total score value and a deviation value for each object OB.
Further, the total value and the deviation value are accumulated in the instruction time column CAT and the deviation value column CAD of the analysis result table TA.
As a result, the difference between the importance level determined by the speaker and the importance level used by the present system is less likely to occur even for materials having complicated object configurations.

Next, the scoring method will be specifically described.
The importance specifying device 220 calculates distances L12 and L32 between the slide instruction position PP2 and the centers C1 and C3 of the circumscribed circle of the object range ROB1, and calculates the ratio of the distances.
Thereafter, the importance specifying device 220 distributes the scores according to the ratio of the distances L12 and L32 so that the object OB having a shorter calculated distance L acquires a higher score.

The data storage unit 300 includes a material database 310 and a time series database 320 which is time series storage means.
Further, the data storage unit 300 is connected to the data collection unit 100, the data totaling unit 200, and the data display unit 400.

The material database 310 is composed of a relational database, for example.
The material database 310 is connected to the analysis device 20, the object specifying device 210, the importance specifying device 220, the importance display device 410 of the data display unit 400, and the condition setting device 420.
The material database 310 includes an analysis result table TA and a slide table TS.

The time series database 320 is constituted by a relational database, for example.
The time series database 320 is connected to the information management device 60, the object specifying device 210, the importance specifying device 220, the condition setting device 420 of the data display unit 400, and the search device 430.
The time series database 320 includes a time series table TT and a score table TSC.

The data display unit 400 includes an importance display device 410 that is an importance display means, a condition setting device 420 that is a condition setting means, a search device 430 that is a search means, and a result display device 440 that is a result display means.
The data display unit 400 is connected to the data storage unit 300.

The importance level display device 410 is constituted by a personal computer, for example, and is connected to the material database 310 and the condition setting device 420.
The importance level display device 410 refers to the material, slide ID, deviation value for each slide, object ID, and deviation value for each object stored in the material database 310.
Thereafter, the deviation values for each material and each slide and each object are visually displayed.

Here, the display of the importance display device will be described with reference to FIG.
FIG. 16 is a configuration diagram of the importance display device.
As shown in FIG. 16, the importance display device 410 includes a display device 411 including a liquid crystal display device, a CRT, and the like, a computer 412 connected to the display device 411, a pointing device 413 connected to the computer 412, and a keyboard 414. Has been.
The display device 411 includes a slide list screen 415 and a slide display screen 416.

The importance level display device 410 acquires the importance level display condition set by the condition setting device 420.
Thereafter, the material indicated by the acquired importance display condition is referred from the material database 310.
The slides included in the referenced material are displayed in a list on the slide list screen 415. Here, the slide list screen 415 displays each slide as a thumbnail on the slide thumbnail SNS.
Thereby, the operator can know the outline of the slide at a glance.

Further, the importance level display device 410 refers to the deviation value for each slide stored in the material database 310.
Thereafter, a predetermined outer frame pattern for the height of the deviation value is added to the outer frame of the slide thumbnail SNS and displayed.
Thereby, the operator can know the importance of the slide at a glance.

Next, the slide display screen 416 will be described.
The importance level display device 410 refers to the deviation value for each object stored in the material database 310.
Thereafter, the slide display screen 416 displays the slide. At that time, a predetermined outer frame pattern for the height of the deviation value is added to the object range ROB and displayed.
Thereby, the operator can know at a glance the importance of the object.

Next, the operation of the importance level display device will be described.
The operator moves the pointer Pt by operating the pointing device 413 or the like. Thereby, a slide to be displayed on the slide display screen 416 is selected from one or a plurality of slide thumbnails SNS displayed on the slide list screen 415.
Thereafter, the importance level display device 410 displays the selected slide on the slide display screen 416.

The condition setting device 420 is constituted by a personal computer, for example, and is connected to the material database 310, the time series database 320, the importance display device, and the search device 430.
The condition setting device 420 sets the display condition of the importance display device 410, the search condition of the search device 430, and the display condition of the result display device 440.

Here, the condition setting device 420 will be described with reference to FIG.
FIG. 17 is a diagram showing the configuration of the condition setting device.
As shown in FIG. 17, the condition setting device 420 includes a display device 421 composed of a liquid crystal display device or CRT, a computer 422 connected to the display device 421, a pointing device 423 connected to the computer 422, and a keyboard 424. ing.

Here, an operation method of the condition setting device will be described.
The operator sets a condition by moving the pointer Pt displayed on the condition setting screen 425 on the display device 421 by operating the pointing device 423 or the like or inputting a condition from the keyboard 424.

When the operator operates the importance button BTD, the condition setting device 420 stores the set condition.
Thereafter, the importance level display device 410 refers to the material ID specified in the above conditions, and displays the slides included therein, the importance level of the slides, and the like.

When the operator operates the search button, the condition setting device 420 similarly stores the set conditions.
Thereafter, the search device 430 refers to the search condition and searches for a scene according to the search condition.
Thereby, the operator can search a scene based on various search conditions.

  Specifically, the condition setting device 420 refers to the material ID, slide ID, object ID, and object type stored in the material database 310 and the lecture ID and speaker ID stored in the time series database 320, and sets the condition setting combo. Display in box CC. Note that “*” is added to the options in consideration of not setting conditions.

As a specific example, a case will be described in which the importance level display device 410 is set to display the importance level for the material identified by the material ID “00001”.
The operator selects “00001” in the material ID combo box CC1. Thereafter, when the importance level button BTD is operated, the condition setting device 420 stores the set condition.
Thereafter, the importance display device 410 refers to the display conditions stored in the condition setting device 420 and displays the importance according to the referred conditions.
The search condition setting operation will be described later.

  The search device 430 is configured by a personal computer, for example, and is connected to the time series database 320, the condition setting device 420, and the result display device 440.

The search device 430 acquires search conditions and display conditions from the condition setting device 420.
After that, scenes that match the search conditions are searched from the time series database 320, and the search results are sorted according to the display conditions.
Further, the sorted search results are transmitted to the result display device 440.

  Here, the operation method of the condition setting device 410 will be described with reference to FIG. 17, and then the search method of the search device 420 will be described with reference to FIGS.

As a specific example, a case where a lecturer searches for a scene indicating a specific object will be described.
In FIG. 17, the operator designates an object ID in the object ID combo box CC6, and designates “*” for the rest. Thereafter, the search button BTS is operated.
Thereby, it is possible to set a desired search condition by operating the condition setting device 420.

Next, a method for the search device 430 to search for a scene in which a speaker has designated an object identified by the object ID “001” will be described with reference to FIG.
FIG. 18 is a diagram for explaining retrieval of a scene in which an object having an object ID “001” is designated.

As illustrated in FIG. 18, the search device 430 searches for records having the stored instruction object ID “001” from the records stored in the time-series table TT of the time-series database 320.
Further, the scene start time TN is acquired from the time information of the retrieved record.
Thus, a scene can be searched based on the object designated by the speaker.

Further, as a specific example, a case will be described in which only a scene in which a speaker indicates a graphic object is searched.
In FIG. 17, the operator selects an object type “F” representing a graphic in the object type combo box CC7, and selects “*” for the rest. Thereafter, the search button BTS is operated.
Thereby, it is possible to set a desired search condition by operating the condition setting device 420.

Next, how the search device 430 searches for a scene in which a lecturer indicates an object whose type is a graphic will be described.
FIG. 19 is a diagram for explaining a scene search method in which an object whose type is a graphic is designated.
As illustrated in FIG. 19, the search device 430 searches for a record in which information indicating that the type of the pointing object stored in the time series table TT included in the time series database 320 is “F” representing a graphic is stored.
Further, the scene start time TN is acquired from the time information of the retrieved record.
Thereby, the operator can search only the scene in which the lecturer indicates the graphic object.

Furthermore, as a specific example, a case will be described in which only scenes presented by a specific speaker are searched.
In FIG. 17, the operator selects a speaker ID in the speaker combo box CC4, and selects “*” for the rest. Thereafter, the search button BTS is operated.
Thereby, it is possible to set a desired search condition by operating the condition setting device 420.

Next, how the search device 430 searches for a scene announced by a specific speaker will be described.
FIG. 20 is a diagram for explaining retrieval of a scene announced by a specific speaker.
As illustrated in FIG. 20, the search device 430 searches for a scene based on the lecturer ID accumulated in the time series table TT included in the time series database 320.
Further, the scene start time TN is acquired from the time information of the retrieved record.
Thereby, a scene can be searched based on a lecturer.

Furthermore, as a specific example, a case will be described in which a scene in which a speaker continues to specify an object (hereinafter referred to as an instruction continuation time) having a length of “5 seconds or longer” is searched.
In FIG. 17, the operator sets “duration” with the radio button CR so that the condition “5 seconds or more” is expressed in the time range designation text boxes CTB1 to CTB3 and the time range designation combo box CC8. Thereafter, the search button BTS is operated.
Thereby, it is possible to set a desired search condition by operating the condition setting device 420.

Next, a description will be given of how to search for a scene whose instruction duration is “5 seconds or longer”.
FIG. 21 is a diagram for explaining a search for a scene having an instruction duration of 5 seconds or more.
As shown in FIG. 21, the search device 430 searches for a record group that has the same instruction object ID accumulated in the time series table TT of the time series database 320 and is continuous when sorted by time information. To do. However, the object with the instruction object ID “000” is excluded.

Thereafter, the search device 430 obtains the instruction duration time by multiplying the number of records in the searched record group by the accumulation time interval.
Furthermore, a scene that matches the search condition for the instruction duration is specified from the obtained instruction duration.
Further, the scene start time TN is acquired from the time information of the retrieved record.
As a result, the operator can search for a scene in which the speaker has instructed a specific object for 5 seconds or longer.

Furthermore, as a specific example, a case will be described in which a scene is searched in which the accumulated time (hereinafter referred to as “instructed accumulated time”) instructed by the speaker is “5 seconds or longer”.
In FIG. 17, the operator sets the condition “cumulative time” with the radio button CR, and “10 seconds or more” with the time range specification text boxes CTB1 to CTB3 and the time range specification combo box CC8. Thereafter, the search button BTS is operated.
Thereby, it is possible to set a desired search condition by operating the condition setting device 420.

Next, a description will be given of how to search for a scene whose instruction accumulation time is “10 seconds or longer”.
FIG. 22 is a diagram for explaining a search for a scene having an instruction accumulation time of 10 seconds or more.
As illustrated in FIG. 22, the search device 430 searches for a record group having the same instruction object ID accumulated in the time series table TT included in the time series database 320. However, the case where the instruction object ID is “000” is excluded.

Thereafter, the search device 430 obtains the instruction accumulated time by multiplying the number of records in the retrieved record group by the accumulation time interval.
Further, a scene that matches the search condition for the instruction cumulative time is specified from the obtained instruction cumulative time.
Further, the scene start time TN is acquired from the time information of the retrieved record.
As a result, the operator can search for a set of scenes in which the lecturer points to a specific object, and the total time length of each scene exceeds 10 seconds.

In addition, as shown in FIG. 17, the operator sets a search condition based on the material ID in the material ID combo box CC1, the conference ID in the conference ID combo boxes CC2 and CC3, and the slide ID in the slide ID combo box CC2. It can be set by the condition setting device 420.
Further, the search device 430 can search based on the material ID, conference ID, and slide ID set by the condition setting device 420.

As a specific example, a case where the result display device 440 displays scenes in ascending order of scene start times will be described.
In FIG. 17, the operator sets a condition “ascending order” in the display method combo box CC9, and sets a search condition other than that.

In particular, when displaying only the third scene in the ascending order, the operator sets the condition “ascending order” in the display method combo box CC9 and the condition “3” in the rank designation text box CTB4. Thereafter, the search button BTS is operated.
As a result, the condition setting device 420 can be operated to set not only the desired search conditions but also the display conditions.

Next, the search device 440 refers to the search conditions and display conditions stored by the condition setting device 420. Thereafter, the scene is searched based on the search condition. Further, the searched scenes are sorted at the start of the scene based on the display conditions.
Thereafter, the sorted search results are transmitted to the search device 430.

Here, when the plurality of search conditions listed above are set at the same time, the search conditions are combined by AND combination.
These can be realized by describing a query or the like using a language such as SQL.

The result display device 440 is composed of a personal computer, for example, and is connected to the search device 430.
The result display device 440 displays the search result received from the 430 search device 430.

Here, the result display device 440 will be described with reference to FIG.
FIG. 23 is a diagram showing the configuration of the result display device.
As shown in FIG. 23, the result display device 440 includes a display device 441 composed of a liquid crystal display device or a CRT, a computer 442 connected to the display device 441, a pointing device 443 connected to the computer 442, and a keyboard 444. ing.
The display device 441 includes a scene display screen 445 and a scene list screen 446.

The scene list screen 446 displays a list of scenes received from the search device 430 and the scene start time TN.
The displayed scene is displayed as a thumbnail in the scene thumbnail SNM.
Thereby, the operator can know at a glance the outline of the searched scene.

The scene display screen 445 displays a scene.
Thereby, the operator can visually recognize the searched scene.
Here, an operation in which the operator selects a scene to be displayed on the scene display screen will be described.

The operator operates the pointing device 443 to move the pointer Pt. Accordingly, a scene to be displayed on the scene display screen 445 is selected from one or a plurality of scene thumbnails SNM displayed on the scene list screen 446.
Thereafter, the result display device 440 displays the selected scene on the scene display screen 445.
Furthermore, the operator can perform playback, stop, fast forward, rewind, pause, frame advance, and the like of the scene by operating the operation button BTM.

In the above-described embodiment, the importance level display device 410, the condition setting device 420, the search device 430, and the result display device 440 have been described as separate devices. However, the present invention is not limited to this, and is realized as one device. Is also possible.
Here, another embodiment in which the data display device 450 has all the functions of the importance display device 410, some functions of the condition setting device 420, and all the functions of the display device 440 will be described with reference to FIG. To do.

FIG. 24 is a configuration diagram of the data display device.
As shown in FIG. 24, the data display device 450 includes a display device 451 such as a liquid crystal display device or a CRT, a computer 452 connected to the display device 451, a pointing device 453 connected to the computer 452, and a keyboard 454. ing.

The display device 451 includes a result display screen 455 and an importance display screen 456.
The result display screen 455 includes a scene display screen 445 and a scene list screen 446.
The importance level display screen 456 includes a slide list screen 415 and a slide display screen 416.

In this embodiment, the operator operates the slide list screen 415, the slide display screen 416, the scene display screen 445, and the scene list screen 446 by the importance display device 410, the condition setting device 420, and the result display device 440, respectively. The same method as described in the embodiment when installed separately is followed.
Also, the functions of the four screens have the same functions as those described in the embodiment when the three devices are installed separately.

Next, a partial function of the condition setting device 420 included in the data display device 450 will be described.
The operator moves the pointer Pt by operating the pointing device 453 or the like. Thereby, the object OB displayed on the slide display screen 416 can be selected.

Thereafter, the data display device 450 changes the search condition stored in the condition setting device 420 so as to search for a scene in which the speaker has designated the object OB selected by the operator.
As a result, the data display device 450 also has some of the functions of the condition setting device 420.

Thereafter, the search device 430 performs a search with reference to the updated search condition stored in the condition setting device 420. Thereafter, the search result is transmitted to the display device 450.
Next, the data display device 450 displays the received search results on the scene display screen 445 and the scene list screen 446.
As a result, the scene instructed by the speaker can be searched for the object OB displayed on the slide display screen 416 selected by the operator.

  In the above-described embodiment, the case where the document is described in the file format for PowerPoint that is software of Microsoft Corporation has been described. However, the present invention is not limited to this. For example, it is also possible to adopt materials described in the file format for Excel or Word, which is the same software of Microsoft Corporation.

In the above embodiment, the display device 10 includes the screen 11, the projector 12, and the slide control device 13, and the case where the pointing position detection device 21 uses the technique of Japanese Patent Application Laid-Open No. 2004-310527 has been described, but the present invention is not limited to this. Absent.
For example, it is possible to employ a combination in which the screen 11 and the projector 12 are devices such as a liquid crystal display, a plasma display, and a CRT display, and the pointing position detection device is a touch panel.
Furthermore, it is also possible to adopt a configuration in which the position of the mouse pointer is acquired as the pointing position by using the pointing position detection device 21 as a personal computer and using Windows (registered trademark) API or the like.

  In the above embodiment, the case where the CCD camera is used has been described. However, the present invention is not limited to this. For example, a camera using a CMOS image sensor or the like may be employed.

  In the above embodiment, the case where the shape of the object range is a square has been described. However, the shape of the object range is not limited to this, and a shape such as a polygon, a circle, an ellipse, a line, or a point can also be adopted.

  In the above embodiment, when a plurality of search conditions are set, the case where each condition is AND-coupled has been described. However, the present invention is not limited to this, and an OR-coupled configuration can also be adopted. .

In the above embodiment, the scoring method based on the ratio of the distance between the slide indication position and the center of the circumscribed circle of the object range has been described as a method of determining the importance when objects overlap each other, but the present invention is not limited to this. Do not mean.
For example, a scoring method based on the ratio of the distance between the slide instruction position and the center and center of gravity of the inscribed circle of the object range and the center of the circumscribed circle of the object itself, the center of the inscribed circle, the center of gravity or the object itself is adopted. Is also possible.

In the above embodiment, the slide list screen 411 has been described with respect to the case where the outer frame pattern predetermined for the importance is added to the slide thumbnail SNS, and the slide display screen 416 is added to the object range ROB for display. However, the present invention is not limited to this.
For example, the outer frame of the slide thumbnail SNS may be displayed in a color predetermined for the importance, and the slide display screen 416 may be configured to display the object range ROB.
In particular, by displaying the importance according to the rule that the deviation value is less than 37.5 in blue, 37.5 to less than 50 in green, 50 to less than 67.5 in yellow, and 67.5 or more in red, It can correspond to a general image of colors for high and low. This makes it easier for the operator to grasp the level of importance.

In the above embodiment, the slide list screen 411 has been described with respect to the case where the outer frame pattern predetermined for the importance is added to the slide thumbnail SNS, and the slide display screen 416 is added to the object range ROB for display. However, the present invention is not limited to this.
It is also possible to fill the slide thumbnail SNS with a semi-transparent color predetermined for the importance, and to fill the object range ROB on the slide display screen 416 in the same manner.

It is a figure which shows the structure which concerns on one Embodiment of this invention, and data transfer between each apparatus. It is a figure for demonstrating the relationship between a document, a slide, an object, and an object range. It is a figure for demonstrating the information which an analyzer has before and after a slide analysis. It is a figure for demonstrating the extraction method of object range information. It is a figure for demonstrating the division | segmentation and joining method of an object. It is the figure which illustrated the correction method of the information regarding an object. It is a block diagram of the analysis result table of a document database. It is a figure for demonstrating the position detection method of a pointing position detection apparatus. It is the figure which illustrated arrangement | positioning of a participant and an instruction | indication apparatus. It is a block diagram of the time series table of a time series database. It is a figure for demonstrating the function of an object specific device. It is a figure for demonstrating the importance specification apparatus. It is a figure for demonstrating the importance specification method for every slide. It is a figure for demonstrating the importance specification method for every object. It is a figure for demonstrating the importance determination method between the mutually overlapping object ranges. It is a block diagram of an importance display apparatus. It is a block diagram of a condition setting device. It is a figure for demonstrating the search method based on instruction | indication object ID. It is a figure for demonstrating the search method based on an instruction | indication object kind. It is a figure for demonstrating the search method based on a lecturer. It is a figure for demonstrating the search method based on instruction | indication continuation time. It is a figure for demonstrating the search method based on instruction | command accumulation time. It is a block diagram of a result display apparatus. It is a block diagram of a data display apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Display apparatus 11 ... Screen 12 ... Projector 13 ... Slide control apparatus 20 ... Analysis apparatus (analysis means)
30 ... Video camera 40 ... Indicating device 41 ... Indicated position detecting device 50 ... Position detecting device (position detecting means)
60 ... Information management device (speaker management means)
DESCRIPTION OF SYMBOLS 100 ... Data collection part 200 ... Data totaling part 210 ... Object identification apparatus (object identification means)
220 ... Importance level specifying device (importance level specifying means)
300 ... Data storage unit 310 ... Material database 320 ... Time series database (time series storage means)
400: Data display unit 410: Importance display device (importance display means)
411 ... display device 412 ... computer 413 ... pointing device 414 ... keyboard 415 ... slide list screen 416 ... slide display screen 420 ... condition setting device (condition setting means)
421 ... Display device 422 ... Computer 423 ... Pointing device 424 ... Keyboard 425 ... Condition setting screen 430 ... Search device (search means)
440 ... Result display device (result display means)
441 ... Display device 442 ... Computer 443 ... Pointing device 444 ... Keyboard 445 ... Scene display screen 446 ... Scene list screen 450 ... Data display device 451 ... Display device 452 ... Computer 453 ... Pointing device 454 ... Keyboard 455 ... Result display screen 456 ... Importance display screen BTD ... Importance buttons BTM1-4 ... Operation buttons BTS ... Search buttons C1-3 ... Center CAD ... Deviation value column CAT ... Cumulative time column CC1-9 ... Condition setting combo box CD ... Document column CKO ... Object type Column CL ... Lecture column CM ... Captured image column CN1-4 ... Corner CO ... Object column CP ... Speaker column CPP ... Instructed position column CR ... Condition setting radio button CRO ... Object range column CS ... Slide column C C1~4 ... score column CT ... Time column CTB1~4 ... conditions set text box
CVF: Vertex coordinates
CVN: Vertex coordinates
D ... Material ID D ... Material ID
IDO1-9 ... Object ID
IDS1-2 ... Slide ID
KOB1-9 ... object types L1-4, L11-L13, L31-L33 ... distance O ... origin OB1-9 ... object
P1-4 ... Participants PP1-3 ... Slide instruction position Pt ... Pointer ROB1-9 ... Object range S1-2 ... Slide SC1-3 ... Scene SNM1-4 ... Scene thumbnail SNS1-4 ... Slide thumbnail T1-3 ... Time information TA ... analysis result table TN1-4 ... scene start time TS ... slide table TSC ... score table TT ... time series table TW1-3 ... scene time length

Claims (16)

Analyzing means for analyzing a slide displayed based on electronic material and extracting an object included in the slide;
Object specifying means for specifying the pointing object instructed by the pointing operation by comparing the object extracted by the analyzing means and pointing position information based on the pointing operation made on the displayed slide;
Time-series storage means for storing the captured image including the displayed slide in association with the pointing object specified by the object specifying means;
An information search system comprising: search means for searching for a desired scene from captured images stored in the time-series storage means based on the instruction object. The analysis means defines information on an object range that is a range in which a speaker determines that the object has been designated,
2. The information search system according to claim 1, wherein the object specifying unit specifies the pointing object from information on the object range determined by the analyzing unit. The information search system according to claim 1, further comprising a result display unit that displays a scene searched by the search unit. 4. The information search system according to claim 1, wherein the result display means displays a thumbnail of the searched scene. Further comprising importance level specifying means for determining the importance level of the slide based on the indication object stored in the time series storage means and time information that is the time indicated by the indication object;
5. The information search system according to claim 1, further comprising importance level display means for visually displaying the importance level of the slide determined by the importance level determination means. 6. The information search system according to claim 5, wherein the importance level display unit displays a thumbnail of the slide. Based on the indication object stored in the time-series storage unit and the time information, further includes an importance level specifying unit that determines the importance level of the object,
5. The information search system according to claim 1, further comprising importance display means for visually displaying the importance of the object determined by the importance determination means. 8. The information search system according to claim 1, further comprising condition setting means for setting search conditions for the search means. The analysis means extracts information on the type of the object,
The object specifying means specifies the type of the pointing object from information on the type of the object extracted by the analyzing means,
9. The information search system according to claim 1, wherein the search means searches for a scene based on the type of the pointing object. The information according to claim 1, wherein the analysis unit adds or deletes the extracted object, and adds, deletes, or changes information on the object range or information on an object type. Search system. 8. The importance level determination unit uses a deviation value of a cumulative value of time when a speaker indicates the slide or the object as an index indicating the level of importance. Information retrieval system. 8. The importance level determination unit uses a deviation value of a scoring result value corresponding to a positional relationship between the indicated position and a displayed object as an index indicating the level of importance. Any information retrieval system described. The information search system according to any one of claims 1 to 12, wherein the search means searches for an important scene based on a duration in which the speaker has indicated the object. The information search system according to any one of claims 1 to 12, wherein the search means searches for a scene based on an accumulated time when the lecturer indicates the object. A speaker management means for storing information for identifying a speaker in the time series storage means;
15. The information search system according to claim 1, wherein the search means searches for a scene based on information for identifying the speaker. Further comprising position detecting means for detecting the position of the pointing device used by the participant and the speaker to point to the displayed slide;
The information retrieval system according to claim 15, wherein the lecturer management unit estimates the presenter from among the participants based on the positions of the participants and the pointing device.

Images