JP2020150519A - Attention degree calculating device, attention degree calculating method and attention degree calculating program - Google Patents

Abstract

To calculate the attention degree of a guest sharing a video captured by a host.SOLUTION: An attention degree calculation server includes an acquisition unit, an identification unit and a calculation unit. The acquisition unit acquires the viewing action data of the guest, based on an input acquired by a guest terminal sharing exhibition video data captured by the host. Based on the viewing action data, the identification unit identifies an object to which the guest pays attention among a plurality of display objects. The calculation unit calculates the attention degree of the plurality of display objects in a manner that the attention degree of the display object, which is identified as an object to which the guest pays attention, relatively increases.SELECTED DRAWING: Figure 3

Description

The present invention relates to a technique for calculating the degree of attention of a video viewer.

Conventionally, a camera capable of live streaming of spherical images capable of viewing 360-degree surroundings mainly by a user has been known. Live streaming of spherical images is expected to be used in various fields such as so-called terrain jigging stance.

Conventionally, a function that allows a plurality of users to share an AR (Augmented Reality) experience has been known. Patent Document 1 discloses a system in which a plurality of users can share MR (Mixed Reality) images and perform collaborative work.

Further, Patent Document 2 discloses that the line of sight of the user wearing the HMD and the point of interest of the user at the tip of the line of sight are specified from the position and orientation of the HMD (Head Mounted Display) in the real space. Has been done.

Further, in Patent Document 3, the degree of attention between the presented images is measured based on the contraction of the pupil of the subject, and the degree of attention is learned as microscopic model information such as a neural network, and this model It describes creating marketing content for viewers with certain attributes based on the information.

Japanese Unexamined Patent Publication No. 2006-293604 Japanese Unexamined Patent Publication No. 2017-33299 Japanese Unexamined Patent Publication No. 2011-118558

An object of the present invention is to calculate the degree of attention of a guest who shares a video taken by a host.

The attention degree calculation device according to the first aspect of the present invention includes an acquisition unit, a specific unit, and a calculation unit. The acquisition unit acquires guest viewing behavior data based on the input acquired at the guest terminal that shares the video data of the exhibition taken by the host. Based on the viewing behavior data, the specific unit identifies the object of interest of the guest among the plurality of exhibition objects at the exhibition. The calculation unit calculates the attention levels of a plurality of exhibition objects so that the attention levels of the exhibition objects specified as the objects of interest of the guests are relatively increased.

Since the attention level calculation device according to the first aspect calculates the attention level of the exhibition target from the guests who virtually participate in the exhibition, the approach is different from the questionnaire to the actual participants of the exhibition. , You can collect data related to your interest in the exhibition target.

In the attention degree calculation device according to the first aspect, the video data may represent a panoramic video. The viewing behavior data may include visual field direction data indicating the visual field direction of the guest at the first reproduction time of the video data, and reproduction time data indicating the first reproduction time. The specific unit extracts the visual field image of the guest at the first playback time from the video data based on the playback time data and the visual field direction data, and identifies the target that the guest paid attention to at the first playback time from the subject of the visual field image. You can. According to this attention degree calculation device, it is possible to extract the visual field image of the guest at the playback time by using the visual field direction data associated with the playback time, and identify the target that the guest has paid attention to from the subject of the visual field image. it can. In general, the field-of-view direction data has a smaller capacity than image data such as a captured image. Therefore, according to this attention level calculation device, the amount of memory used and the amount of data communication related to the calculation of the degree of attention can be saved. There is a possibility that it can be done.

In the attention degree calculation device according to the first aspect, the video data may represent a panoramic video. The viewing behavior data includes playback time data indicating the first playback time of the corresponding video data when the guest terminal captures the video data, and viewing direction data indicating the viewing direction of the guest at the first playback time. It's fine. The specific unit extracts the visual field image of the guest at the first playback time from the video data based on the playback time data and the visual field direction data, and identifies the target that the guest paid attention to at the first playback time from the subject of the visual field image. You can. According to this attention calculation device, the visual field image of the guest at the playback time is extracted by using the visual field direction data associated with the playback time when the guest performs the capture operation, and the visual field image of the guest is extracted from the subject of the visual field image. It is possible to identify the target that the guest has paid attention to. Since it can be estimated that the visual field image of the guest when the guest performs the capture operation is likely to include the object that the guest paid attention to, according to this attention level calculation device, a highly reliable attention level can be obtained. Can be calculated. In general, the field-of-view direction data has a smaller capacity than image data such as a captured image. Therefore, according to this attention level calculation device, the amount of memory used and the amount of data communication related to the calculation of the degree of attention can be saved. There is a possibility that it can be done.

In the attention degree calculation device according to the first aspect, the viewing behavior data is the captured image of the video data generated by the guest terminal in response to the capture operation, and the first of the video data corresponding to the capture operation when the capture operation is performed. It may include playback time data indicating the playback time. The specific unit may specify the target that the guest paid attention to at the first reproduction time from the subject of the captured image. According to this attention degree calculation device, it is possible to identify the target that the guest has paid attention to from the subject of the captured image by the guest. Since it can be estimated that such a captured image is likely to include an object that the guest has paid attention to, the attention degree calculation device can calculate a highly reliable attention degree.

In the attention degree calculation device according to the first aspect, the viewing behavior data is a GUI (Graphical User Interface) component displayed superimposed on the video represented by the video data on the guest terminal at the first playback time of the video data. Among them, GUI selection data indicating the first GUI component selected by the guest and reproduction time data indicating the first reproduction time may be included. The specific unit may extract the visual field image of the guest from the video data based on at least the reproduction time data, and specify the target of the guest's attention at the first reproduction time based on the GUI selection data from the subject of the visual field image. According to this attention degree calculation device, the visual field image of the guest can be extracted based on at least the reproduction time data, and the target that the guest pays attention to from the subject of the visual field image can be specified based on the GUI selection data. The GUI selection data shows the GUI parts actively selected by the guest, and is the guest really interested in some or all of the exhibition objects included in the visual field image from the selected GUI parts? Or you can estimate whether you want to see another exhibit sooner. That is, in this way, by specifying the target that the guest has paid attention to in consideration of the GUI selection data, it is possible to calculate the degree of attention with high reliability.

In the attention degree calculation device according to the first aspect, the viewing behavior data includes the visual field direction data indicating the visual field direction of the guest at the first reproduction time of the video data and the reproduction time data indicating the first reproduction time. It's fine. The specific unit determines the visual field range of the guest at the first playback time based on the reference position / orientation data and the visual field direction data indicating the position of the host at the first playback time and the reference angle of the video data, and determines the visual field. The object of interest of the guest at the first playback time may be specified based on the range and the map data that associates each of the plurality of exhibition objects with the position of the exhibition object.

According to this attention degree calculation device, the visual field direction data associated with the reproduction time and the reference position / orientation data associated with the same reproduction time are used to determine the visual field range of the guest at the reproduction time, and this visual field. It is possible to identify the target that the guest has focused on based on the range and the map data. In general, the reference position / orientation data and the map data have a smaller capacity than the video data, so that the attention level calculation device may save the memory usage related to the attention level calculation. ..

In the attention degree calculation device according to the first aspect, the video data may represent a panoramic video. The viewing behavior data includes playback time data indicating the first playback time of the corresponding video data when the guest terminal captures the video data, and viewing direction data indicating the viewing direction of the guest at the first playback time. It's fine. The specific unit determines the visual field range of the guest at the first playback time based on the reference position / orientation data and the visual field direction data indicating the position of the host at the first playback time and the reference angle of the video data, and determines the visual field. The object of interest of the guest at the first playback time may be specified based on the range and the map data that associates each of the plurality of exhibition objects with the position of the exhibition object.

According to this attention calculation device, the guest at the playback time is performed by using the visual field direction data associated with the playback time when the guest performs the capture operation and the reference position / orientation data associated with the same playback time. It is possible to determine the field of view of the guest and identify the target that the guest has focused on based on this field of view and the map data. Since it can be estimated that the visual field range of the guest when the guest performs the capture operation is likely to include the object that the guest has paid attention to, according to this attention level calculation device, a highly reliable attention level can be obtained. Can be calculated. In general, the reference position / orientation data and the map data have a smaller capacity than the video data, so that the attention level calculation device may save the memory usage related to the attention level calculation. ..

In the attention level calculation device according to the first aspect, the calculation unit increases the attention level of the exhibition object specified as the guest's attention target according to the length of the duration of the guest's attention to the exhibition target. You may calculate the degree of attention of multiple exhibition objects. According to this attention calculation device, the longer the guest gazes at the exhibition object, the more attention the exhibition object has, so that the exhibition object and the guest's eyes are only noticed by the guest for a short time. It is possible to calculate the degree of attention by properly distinguishing it from the exhibition target that has stolen for a long time.

The method of calculating the degree of attention according to the second aspect of the present invention is that the computer acquires the guest viewing behavior data based on the input acquired at the guest terminal sharing the video data of the exhibition taken by the host. , The computer identifies the guest's attention among multiple exhibition objects at the exhibition based on the viewing behavior data, and the computer identifies the guest's attention as the guest's attention. It includes calculating the degree of attention of a plurality of exhibition objects so as to increase the number of objects.

Since this attention level calculation method calculates the attention level of the exhibition target from the guests who virtually participate in the exhibition, the approach to the exhibition target is different from the questionnaire to the actual participants of the exhibition. It is possible to collect data related to interests.

The attention degree calculation program according to the third aspect of the present invention is a means for acquiring guest viewing behavior data based on the input acquired by the host at the guest terminal sharing the video data of the exhibition taken by the host. Based on the viewing behavior data, a means for identifying a guest's attention target among a plurality of exhibition targets in an exhibition, and a plurality of exhibition objects so that the attention level of the exhibition target specified as the guest's attention target increases relatively. It functions as a means to calculate the degree of attention of the exhibition target.

Since this attention level calculation program calculates the attention level of the exhibition target from the guests who virtually participate in the exhibition, the approach to the exhibition target is different from the questionnaire to the actual participants of the exhibition. It is possible to collect data related to interests.

According to the present invention, it is possible to calculate the degree of attention of a guest who shares a video shot by a host.

The block diagram which illustrates the video sharing system which includes the attention degree calculation server which concerns on embodiment. The block diagram which illustrates the guest terminal in FIG. The block diagram which illustrates the attention degree calculation server which concerns on embodiment. Explanatory view of the guest's field of view. The figure which illustrates the visual field image of a guest. Explanatory drawing of video partition. The figure which illustrates the GUI component which is superposed on the image and is displayed. Explanatory drawing of attention degree calculation using map data. The flowchart which illustrates the operation of the attention degree calculation server of FIG.

Hereinafter, embodiments will be described with reference to the drawings. Hereinafter, elements that are the same as or similar to the elements described will be designated by the same or similar reference numerals, and duplicate explanations will be basically omitted. For example, when a plurality of the same or similar elements exist, a common code may be used to explain each element without distinction, and the common code may be used to explain each element separately. In addition, the branch number may be used.

(Embodiment)
In the attention degree calculation server according to the embodiment, the guest terminal (hereinafter referred to as the guest terminal) shares the video data shot and posted by the host terminal (hereinafter referred to as the host terminal) via the video distribution server. Can be incorporated into a video sharing system. FIG. 1 illustrates a video sharing system including the attention level calculation server 40 according to the embodiment.

The video sharing system of FIG. 1 includes a host terminal 10, a video distribution server 20, guest terminals 30-1, 30-2, ..., And a attention calculation server 40. The video distribution server 20 of FIG. 1 is connected to a part or all of the host terminal 10, the guest terminal 30, and the attention calculation server 40 via a network such as the Internet, and can transmit and receive data to and from each other. Further, the attention degree calculation server 40 is connected to a part or all of the host terminal 10, the video distribution server 20, and the guest terminal 30 via a network, and can transmit and receive data to and from each other.

The host terminal 10, the video distribution server 20, and the guest terminal 30 have basic functions for exchanging and receiving video sharing services, as described below. In the following description, it is assumed that the video sharing system of FIG. 1 shares live video, that is, live streaming, but attention will be described later not only when sharing live video but also when sharing video created in advance by the host. It is possible to calculate the degree. Therefore, the embodiment can be applied not only to live video but also to video in general. In addition, the video is premised on the shooting of the exhibition by the host. That is, the guest will share the video data of the host shooting the exhibition. Here, an exhibition is typically a trade fair for the industry, but is not limited to artificial objects (for example, products, works, buildings, (virtual) objects, etc.) and natural products (landscapes, humans). It can mean a place where you can appreciate the exhibits such as singing, dancing, playing, sports, etc., or a combination of these. Specifically, the exhibition may include concerts, plays, sports competitions, and the like.

The host terminal 10 sequentially encodes live video / audio data generated by, for example, a camera / microphone connected to the host terminal 10, and transmits (uploads) the encoded (video or multimedia) data to the video distribution server 20. ). Here, the camera may be a 360-degree spherical or hemispherical 360-degree camera or other panoramic camera, or may be a normal monocular camera.

The host terminal 10 may be an electronic device such as a computer, for example, a mobile terminal (for example, a smartphone, a tablet, a laptop, etc.), a PC (Personal Computer), or the like, or a VR (Virtual Reality) having a communication function. ) / AR / MR terminals, such as HMDs, but are not limited to these. When a head-mounted device such as an HMD is used as the host terminal 10, the field of view of the host can be secured to some extent by adopting a monocular or transmissive device.

The video distribution server 20 receives the encoded data from the host terminal 10, and performs processing such as re-encoding on the encoded data as necessary. The video distribution server 20 distributes the (processed) encoded data to the guest terminal 30.

The guest terminal 30 receives the encoded data from the video distribution server 20, decodes the encoded data, and reproduces the live video (which may include audio). When reproducing a panoramic image including a 360-degree image, the guest terminal 30 detects, for example, the direction of the guest's head and / or the line of sight, or the direction of the guest terminal 30, and reproduces the image based on this. You may switch the viewpoint of the image. Similarly, when playing back a panoramic image, the guest terminal 30 detects, for example, an operation for switching the viewpoint by the guest, for example, flick, stroke, touch, tap, swipe, drag, click, etc. You may switch the viewpoint of the image to be reproduced based on.

The guest terminal 30 is typically a combination of a display device such as a VR / AR / MR terminal such as an HMD having an excellent immersive feeling and this control device, for example, a mobile terminal, a PC, a game machine, or the like. Good. The display device and its control device do not have to be separate types and may be integrated.

The guest can experience as if he / she is participating in the exhibition through the image taken by the host in a remote place and reproduced by the guest terminal 30. Then, the attention degree calculation server 40 according to the embodiment collects guest viewing behavior data based on the input acquired by the guest terminal 30 at the time of video reproduction, and based on this viewing behavior data, a plurality of exhibition targets at the exhibition. Identify the target of the guest's attention. Further, the attention level calculation server 40 calculates the attention level of these exhibition objects based on the objects of interest of the guest.

In this way, the attention level calculation server 40 calculates the attention level from the guests who virtually participate in the exhibition, so that the interest in the exhibition target is different from the questionnaire to the actual participants of the exhibition.・ You can collect data related to your interests. It should be noted that this degree of attention can be expected to be used in various ways in combination with or in place of the analysis results of conventional questionnaires. For example, attention is most photographed by guests, providing feedback information to exhibitors such as companies exhibiting at exhibitions or exhibition organizers, supporting marketing to guests or other (potential) consumers, etc. It can be used for requesting the host to shoot the exhibition target (shooting target) that would be desired. As an example, in a concert where multiple people watch singing, a ranking of these people is created based on the degree of attention of the guest, and a person who is strongly interested by the individual guest is the degree of attention of the guest. It can also be specified based on. In addition, as an example of utilizing such attention, a message from a person who is strongly interested by an individual guest may be sent to the guest.

The number of each device shown in FIG. 1 is merely an example. For example, the number of guest terminals 30 changes from moment to moment, so that it may be 0, or it may be hundreds or thousands. Further, since the video sharing system of FIG. 1 can deliver a plurality of live videos in parallel, the number of host terminals 10 can be two or more.

Hereinafter, the configuration of the guest terminal 30 will be described in detail. As illustrated in FIG. 2, the guest terminal 30 includes a communication I / F 100, a processor 110, a memory 120, an output device 130, and an input device 140.

The communication I / F 100 is a module for communicating with an external device such as a video distribution server 20, a attention calculation server 40, etc. via a network, and is, for example, a signal processing circuit for transmission / reception, an antenna. , LAN (Local Area Network) terminal and the like may be included. The communication I / F can be, for example, a module for wide area communication such as mobile communication, a module for wireless / wired LAN, a module for Bluetooth (registered trademark), and the like.

The processor 110 is typically a CPU (Central Processing Unit) and / or a GPU (Graphics Processing Unit), but is a microcomputer, FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), or other general-purpose processor. It may be a processor of. The processor 110 performs, for example, input / output control, communication control, image / audio processing, and generation of viewing behavior data described later.

The memory 120 temporarily stores a program executed by the processor 110 for the processor 110 to realize each process and data used by the processor 110, for example, input data, video data, viewing behavior data, and the like. To do. The memory 120 may include a RAM (Random Access Memory) having a work area in which such programs / data are deployed.

The output device 130 may include, for example, a display device that outputs video reproduced by the processor, a speaker that outputs audio reproduced by the processor, and the like. The output device 130 may be a part or all of the elements of the HMD.

The input device 140 may include, for example, some or all of a sensor, touch screen, microphone, keyboard, mouse, and the like.

Here, the sensor may include motion sensors such as accelerometers, gyroscopes, magnetic sensors, and the like. The sensor data detected by the motion sensor can be used to estimate the direction (for example, the inclination of three axes) of the guest's head and / or the guest terminal 30. Specifically, at the time of reproducing the panoramic image, the Yaw angle, the Roll angle, and the Pitch angle, which are the three-dimensional rotation angles of the guest head and / or the guest terminal 30, are estimated based on the sensor data. The viewpoint of the image displayed on the guest terminal 30 can be controlled accordingly.

The sensor may also include a so-called eye tracking device. The eye tracking device captures the movement of the guest's eyeball and generates input data indicating the direction of the guest's line of sight. Such input data can also be used to control the viewpoint of the image displayed on the guest terminal 30.

By executing the program stored in the memory 120, the processor 110 executes the received data acquisition unit 111, the decoder 112, the video output unit 113, the input data acquisition unit 114, the viewing behavior data generation unit 115, and the viewing behavior in FIG. It can function as a reporting unit 116. Further, the memory 120 may include the video data storage unit 121 of FIG. 2, the viewing behavior data storage unit 122, and the input data storage unit 123.

The communication I / F 100 includes a receiving unit 101 that receives various data from an external device via a network, and a transmitting unit 102 that transmits various data to the external device via the network. The receiving unit 101 receives the above-mentioned encoded data from the video distribution server 20 and sends it to the receiving data acquisition unit 111. Further, the transmission unit 102 receives the viewing behavior data from the viewing behavior reporting unit 116 and transmits the viewing behavior data to the attention degree calculation server 40.

The reception data acquisition unit 111 acquires reception data from the reception unit 101 and writes it in the memory 120. As an example, the reception data acquisition unit 111 acquires encoded data from the reception unit 101 and writes it in the video data storage unit 121.

The video data storage unit 121 stores the encoded data acquired by the reception data acquisition unit 111. This encoded data includes at least encoded video data, and may further include encoded audio data. The encoded video data stored in the video data storage unit 121 can be read by various functional units of the processor 110, for example, the decoder 112.

The decoder 112 reads the encoded video data from the video data storage unit 121 from the video data storage unit 121. The decoder 112 decodes the encoded video data according to an appropriate codec and reproduces the video data. The decoder 112 may include an audio decoder in addition to the video decoder. The decoder 112 sends the reproduced video data to the video output unit 113. Further, although the decoder 112 is drawn as if it is a part of the function of the processor 110 in FIG. 2, it may be a dedicated circuit.

The video output unit 113 receives the video data from the decoder 112 in a timely manner, that is, before the arrival of the output timing corresponding to the video data, and sends the video data to the output device 130. The video output unit 113 may control the output of audio data as well as video data.

When the video data represents a panoramic video, the video output unit 113 may control the viewpoint of the video to be reproduced. For example, the video output unit 113 is for input data representing the direction of the guest's head and / or line of sight stored in the input data storage unit 123, or the posture of the guest terminal 30, and / or for the guest to switch the viewpoint. The viewpoint of the image to be reproduced can be determined based on the input data representing the operation. Then, the video output unit 113 may generate video data according to the determined viewpoint based on the video data from the decoder 112, and send this to the output device 130.

In addition, the video output unit 113 displays GUI (Graphical User Interface) components and / or partitions to be described later (for example, partition boundaries and / or partition labels) as necessary for the video data. May be added and then sent to the output device 130.

The output device 130 receives video data from the video output unit 113 and displays the video data. The output device 130 may output audio data in addition to displaying video data.

The input device 140 generates input data based on the input in the guest terminal 30, and sends the input data to the input data acquisition unit 114. Here, the input data is, for example, data representing the direction of the guest's head and / or the line of sight, or the direction of the guest terminal 30, and the guest's operation on the display image of the guest terminal 30 (for example, for capture operation and viewpoint switching). Operation, operation to select GUI parts superimposed on the video, data representing the video partition), etc., and playback showing the playback time of the corresponding video data when such data is generated. It may include time data.

The guest terminal 30 may optionally implement a capture function. When the guest terminal 30 implements the capture function, the processor 110 captures the image displayed on the display device included in the output device 130 and generates the captured image according to the input data representing the capture operation, for example. Can function as a capture unit (not shown).

The input data acquisition unit 114 receives various input data from the input device 140 and writes them in the input data storage unit 123.

The input data storage unit 123 stores the input data acquired by the input data acquisition unit 114. The input data stored in the input data storage unit 123 is read out by various functional units of the processor 110, for example, the viewing behavior data generation unit 115.

The viewing behavior data generation unit 115 reads the input data from the input data storage unit 123, and based on this, generates viewing behavior data representing the conscious or unconscious behavior of the guest when viewing the video. The viewing behavior data generation unit 115 writes the generated viewing behavior data in the viewing behavior data storage unit 122.

Specifically, when the video data is a panoramic video, that is, a video whose viewpoint can be switched, the viewing behavior data generation unit 115 corresponds to the visual field direction data representing the visual field direction of the guest at a certain time point and the time point. The viewing behavior data including the reproduction time data indicating the reproduction time of the video data to be performed may be generated. Here, the certain time point may be, for example, a predetermined timing, or may be, for example, a timing when the guest performs a predetermined input such as a capture operation.

The viewing behavior data generation unit 115 can generate viewing direction data based on data representing the direction of the guest's head and / or the line of sight, or the direction of the guest terminal 30, which can be included in the input data, for example. As an example, the viewing behavior data generation unit 115 views the guest's field of view in the front direction (F), the right direction (R), the left direction (L), and the rear direction (L) with respect to the host's field of view, as illustrated in FIG. It is divided into four visual field regions of B), it is determined which visual field region the direction of the guest's head and / or the line of sight, or the direction of the guest terminal 30 corresponds to, and the index representing the determined visual field region is the visual field. It may be generated as direction data. Note that FIG. 4 is merely an example, and the guest's visual field may be divided into a visual field region of 3 or less, or may be divided into a visual field region of 5 or more. Further, the visual field region may or may not coincide with the unit for switching the viewpoint of the image.

As will be described later, by combining the visual field direction data and the reproduction time data with the video data, the visual field image of the guest at the reproduction time indicated by the reproduction time data can be reproduced to some extent. This field image may contain the guest's attention. For example, the field image illustrated in FIG. 5 includes booth 301, exhibits 301-1, 301-2, 301-3, booth 302, and exhibit 302-1 as subjects, but the guest is one of these. You may be interested in some or all of them.

Further, when the video data is a panoramic video or a normal video, that is, a video whose viewpoint cannot be switched, the viewing behavior data generation unit 115 includes the captured image generated in response to the capture operation by the guest and the captured image. Viewing behavior data including playback time data indicating the playback time of the corresponding video data when the capture operation is performed may be generated. Although the captured image has a larger data capacity than the visual field direction data, it is the visual field image itself of the guest when the capture operation is performed, so that the process of reproducing the visual field image is unnecessary and the visual field caused by such processing. There is an advantage that the error does not occur. In addition, it can be estimated that the captured image saved by the guest actively performing the capture operation is likely to include the object of interest of the guest.

The video may be divided into a plurality of partitions and displayed, for example, as shown in FIG. In the example of FIG. 6, the image is divided into three partitions L (left), C (center), and R (right). If the guest performs an operation to select one of the partitions, it is highly probable that the partition contains the object of interest of the guest. For example, in the example of FIG. 6, if the guest selects partition L, it can be estimated that the guest is interested in the exhibit 301-1 and / or booth 301, and if the guest selects partition C, the guest can be estimated to be interested. Can be presumed to be interested in exhibit 301-2, exhibit 301-2 and / or booth 301, and if the guest selects partition R, the guest will be in exhibit 302-1 and / or booth 302. It can be estimated that you are interested. However, FIG. 6 is merely an example, and the video may be divided into 2 or less partitions or 4 or more partitions. Further, for example, a partition corresponding to each object may be determined by using an object recognition technique or the like.

Therefore, the viewing behavior data generation unit 115 corresponds to the viewing direction data and / or the captured image when the partition is selected by the guest, the partition selection data indicating the selected partition, and when the partition is selected. Viewing behavior data including playback time data indicating the playback time of the video data may be generated.

Further, GUI components may be superimposed and displayed on the video, for example, as shown in FIG. The guest who selected the GUI component 401 in FIG. 7 is interested in at least a part of the booth 301, which is the subject of the currently displayed image, and the exhibits 301-1, 301-2, and 301-3. Can be estimated. On the other hand, the guest who selected the GUI component 402 in FIG. 7 is interested in at least a part of the booth 301, which is the subject of the image currently displayed, and the exhibits 301-1, 301-2, and 301-3. It can be presumed that there is no or that you are interested in other booths and / or exhibits.

Therefore, the viewing behavior data generation unit 115 indicates the GUI selection data indicating which of the GUI parts displayed superimposed on the video is selected by the guest, and the playback time of the corresponding video data when such selection is made. Viewing behavior data including playback time data may be generated. When the video data is a panoramic image, the viewing behavior data generation unit 115 selects the visual field direction data and / or the captured image when the GUI component is selected so that the visual field image of the guest can be reproduced. Must be included in viewing behavior data.

The viewing behavior data storage unit 122 stores the viewing behavior data generated by the viewing behavior data generation unit 115. The viewing behavior data stored in the viewing behavior data storage unit 122 is read out by various functional units of the processor 110, for example, the viewing behavior reporting unit 116.

The viewing behavior reporting unit 116 reads the viewing behavior data from the viewing behavior data storage unit 122 and reports this to the attention level calculation server 40. Specifically, the viewing behavior reporting unit 116 sends the viewing behavior data to the transmitting unit 102, and the transmitting unit 102 transmits the viewing behavior data to the attention degree calculation server 40 via the network.

Next, the configuration of the attention degree calculation server 40 will be described in detail. The attention calculation server 40 includes a communication I / F 200, a processor 210, and a memory 220, as illustrated in FIG.

The communication I / F 200 is a module for communicating with an external device such as a host terminal 10, a video distribution server 20, a guest terminal 30, etc. via a network, and is, for example, a signal processing circuit for transmission / reception. , Optical connectors, etc. may be included. The communication I / F 200 can be, for example, an optical communication module.

The processor 210 may be a general-purpose or dedicated processor or the like. The processor 210 performs, for example, input / output control, communication control, identification of an object of interest and calculation of the degree of attention, which will be described later.

The memory 220 temporarily stores a program executed by the processor 210 and data used by the processor 210 for the processor 210 to realize each process, such as viewing behavior data, attention target data, attention level, and the like. Store. Further, the memory 220 may store video data and / or store reference position / orientation data and map data, which will be described later, in order to reproduce the visual field image of the guest associated with the viewing behavior data. The memory 220 may include a RAM having a work area in which such programs / data are deployed.

The processor 210 can function as the received data acquisition unit 211, the attention target identification unit 212, the attention degree calculation unit 213, and the attention degree reporting unit 214 of FIG. 3 by executing the program stored in the memory 220. Further, the memory 220 may include the viewing behavior data storage unit 221, the attention level storage unit 222, and the attention target data storage unit 223 in FIG.

The communication I / F 200 includes a receiving unit 201 that receives various data from an external device via a network, and a transmitting unit 202 that transmits various data to the external device via the network. The reception unit 201 receives the viewing behavior data from the guest terminal 30 and sends it to the reception data acquisition unit 211. The reception unit 201 may receive (encoded) video data and / or reference position / orientation data from the host terminal 10 and / or the video distribution server 20 and send the data to the reception data acquisition unit 211. Here, the reference position / orientation data indicates the position of the host at a given time and the reference angle of the video data. Further, the transmission unit 202 receives the attention level from the attention level reporting unit 214 and transmits this to, for example, an attention level utilization server (not shown).

Here, the attention level utilization server receives the attention level from the attention level calculation server 40 via the network and utilizes it. Specifically, the attention level utilization server provides feedback information created based on the attention level to the exhibitor of the exhibition or the organizer of the exhibition, or to guests or other (potential) consumers. It supports marketing based on the degree of attention, determines the shooting target based on the degree of attention, and requests the host to shoot the shooting target.

The reception data acquisition unit 211 acquires the reception data from the reception unit 201 and writes it in the memory 220. As an example, the reception data acquisition unit 211 acquires the viewing behavior data from the receiving unit 201 and writes it in the viewing behavior data storage unit 221. Further, the reception data acquisition unit 211 can acquire (encoded) video data and / or reference position / orientation data from the reception unit 201 and write them to the memory 120.

The viewing behavior data storage unit 221 stores the viewing behavior data acquired by the received data acquisition unit 211. The viewing behavior data stored in the viewing behavior data storage unit 221 can be read out by various functional units of the processor 210, for example, the attention target identification unit 212.

The viewing behavior data may be collected without distinguishing between the host and the guest, or may be collected by distinguishing between the host and the guest. That is, the viewing behavior data generated when viewing the video transmitted by one host terminal 10 may be distinguished from the viewing behavior data generated when viewing the video transmitted by another host terminal 10. In this case, the viewing behavior data may include a host identifier indicating which host terminal 10 is the source of the video and the viewing behavior data was generated at the time of viewing. Further, the viewing behavior data generated by the guest terminal 30 of a certain guest may be distinguished from the viewing behavior data generated by the guest terminal 30 of another guest. In this case, the viewing behavior data may include a guest identifier indicating which guest terminal 30 generated the viewing behavior data.

The attention target identification unit 212 reads the viewing behavior data from the viewing behavior data storage unit 221 and reads the reference data of a plurality of exhibition targets in the exhibition from the memory 220. Here, the reference data may include, for example, the appearance of the exhibition object (which may include a partial appearance such as the logo and / or trademark of the exhibition object and / or the exhibitor to which the exhibition object belongs, the face of a person as the exhibition object, and the like). It can indicate the code given to the object (eg, one-dimensional code, two-dimensional code, etc.), the text associated with the exhibit (eg, name, nickname, identification number, description, etc.), the location of the exhibit, etc. .. Reference data indicating the positions of a plurality of exhibition objects can also be called map data. The attention target identification unit 212 identifies the target that the guest has paid attention to at the playback time of the video data indicated by the playback time data included in the viewing behavior data based on the viewing behavior data and the reference data. Then, the attention target identification unit 212 generates, for example, the attention target data including the reproduction time data and the exhibition target identifier that identifies the specified target, and writes this in the attention target data storage unit 223.

When the viewing behavior data includes the above-mentioned host identifier and / or guest identifier, the attention target data may also include this. Further, the host identifier and / or the guest identifier may be converted into an attribute identifier that identifies one or more attributes, for example, gender, age, preference, skill, and the like. The conversion rule from the host identifier and / or the guest identifier to the attribute identifier may be created by, for example, the server that manages the user information of the video sharing system of FIG. 1, and the attention degree calculation server 40 may acquire this. For example, a guest identifier may be converted into an attribute identifier that identifies attributes such as "male", "30's", "early adapter", and "expert" and included in the data of interest.

Specifically, the attention target identification unit 212 can specify the target that the guest has paid attention to by various techniques as listed below.

(1) It is assumed that the viewing behavior data includes the above-mentioned visual field direction data and playback time data, and the video data (representing a panoramic image) is stored in the memory 220. In this case, the attention target identification unit 212 extracts the visual field image of the guest at the reproduction time indicated by the reproduction time data from the video data based on the reproduction time data and the visual field direction data. For example, the attention target identification unit 212 extracts one panoramic image corresponding to the reproduction time indicated by the reproduction time data from the video data, and further extracts one viewpoint corresponding to the visual field direction indicated by the visual field direction data from this panoramic image. Image can be extracted as a guest field image.

Then, the attention target identification unit 212 identifies the target that the guest has paid attention to at the reproduction time indicated by the reproduction time data from the subject of the visual field image. Specifically, the attention target identification unit 212 compares the appearance of the subject with the appearance of a plurality of exhibition objects included in the reference data, extracts the text area or the code area included in the subject, and refers to the extracted area. Determine whether the subject included in the field image corresponds to any of the multiple exhibition objects by comparing it with the text associated with multiple exhibition objects included in the data or the code assigned to the exhibition object. it can.

(2) It is assumed that the viewing behavior data includes the above-mentioned captured image and playback time data. In this case, the attention target identification unit 212 identifies the target that the guest has paid attention to at the reproduction time indicated by the reproduction time data from the subject of the captured image. Specifically, the attention target identification unit 212 can determine whether or not the subject included in the captured image corresponds to any of a plurality of exhibition targets by the same technique as in (1) above.

(3) It is assumed that the viewing behavior data includes the above-mentioned GUI selection data and playback time data. If the viewing behavior data further includes a captured image, or if the video data (representing a normal video) is stored in the memory 220, the attention target identification unit 212 may use the captured image or video. Of the data, one image corresponding to the reproduction time indicated by the reproduction time data can be treated as a guest field image. On the other hand, when the viewing behavior data further includes the visual field direction data and the video data (representing the panoramic image) is stored in the memory 220, the attention target identification unit 212 is the same as the above (1). The technique allows the guest's visual field image to be extracted.

Then, the attention target identification unit 212 extracts the subject corresponding to any of the plurality of exhibition objects from the subject of the field image by the same technique as in (1) above, and further extracts the extracted exhibition object based on the GUI selection data. By narrowing down, the target that the guest paid attention to at the playback time indicated by the playback time data is specified. For example, when the GUI selection data indicates that a GUI component such as "exhibit on the right" has been selected, the target identification unit 212 pays attention to the exhibit located on the right side in the visual field image. Can be specified as. Further, when the GUI selection data indicates that, for example, the GUI component 401 of FIG. 7 has been selected, the attention target identification unit 212 can specify the exhibition target included in the visual field image as the target that the guest has paid attention to. On the other hand, when the GUI selection data indicates that, for example, the GUI component 402 of FIG. 7 has been selected, the attention target identification unit 212 may exclude the exhibition target included in the visual field image from the target paid by the guest.

(4) It is assumed that the viewing behavior data includes the partition selection data and the playback time data described above. If the viewing behavior data further includes a captured image, or if the video data (representing a normal video) is stored in the memory 220, the attention target identification unit 212 may use the captured image or video. Of the data, one image corresponding to the reproduction time indicated by the reproduction time data can be treated as a guest field image. On the other hand, when the viewing behavior data further includes the visual field direction data and the video data (representing the panoramic image) is stored in the memory 220, the attention target identification unit 212 is the same as the above (1). The technique allows the guest's visual field image to be extracted.

Then, the attention target identification unit 212 further narrows down this field of view image based on the partition selection data. For example, when the partition selection data indicates that the partition L of FIG. 6 is selected, for example, the attention target identification unit 212 may trim the portion corresponding to the partition L from the visual field image. Then, the attention target identification unit 212 can determine whether or not the subject included in the narrowed-down visual field image corresponds to any of the plurality of exhibition targets by the same technique as in (1) above.

(5) It is assumed that the viewing behavior data includes the above-mentioned visual field direction data and reproduction time data, and the map data and the reference position / orientation data are stored in the memory 220. The attention target identification unit 212 determines the position of the host at the reproduction time indicated by the reproduction time data and the reference angle of the video data based on the reproduction time data and the reference position / orientation data. Further, the attention target identification unit 212 rotates the determined reference angle of the video data according to the visual field direction indicated by the visual field direction data, and estimates the visual field range of the guest at the reproduction time indicated by the reproduction time data. Map data associates multiple exhibits in an exhibition with their location. Then, the attention target identification unit 212 identifies the exhibition target displayed within the estimated guest's visual field range as the target that the guest has paid attention to at the reproduction time indicated by the reproduction time data. Specifically, as illustrated in FIG. 8, the attention target identification unit 212 estimates the guest field of view 502 centered on the host position 501, and the booth A01 exhibited in this field of view 502. , And / or the exhibits in the booth A01 can be identified as objects of interest to the guest.

According to the technique of (5) above, the amount of memory 220 used is reduced as compared with the technique of storing a large amount of video data in the memory 220 as described in (1), (3) to (4) above. There is a possibility of saving. In addition, according to the technique (5) above, the amount of memory 220 used can be further saved by using the same map data when calculating the degree of attention related to the images of other hosts participating in the same exhibition. .. Alternatively, in the techniques (1), (3) to (4) above, the attention level calculation server 40 sends an image corresponding to the playback time indicated by the playback time data among the video data to the host terminal 10 or the video distribution server 20. You may request it.

The attention target data storage unit 223 stores the attention target data generated by the attention target identification unit 212. The attention target data stored in the attention target data storage unit 223 is read out by various functional units of the processor 210, for example, the attention degree calculation unit 213. Note that the data of interest may include, for example, a playback time, a guest identifier, a host identifier, and / or an attribute identifier as a key, and data that matches a desired key may be selectively readable. ..

The attention level calculation unit 213 reads the attention target data from the attention target data storage unit 223, and calculates the attention levels of a plurality of exhibition targets based on the read data. Then, the attention degree calculation unit 213 writes the calculated attention degree to the attention degree storage unit 222.

The attention level calculation unit 213 can calculate the attention level of a plurality of exhibition objects so that the attention level of the object indicated by the attention target data is relatively increased. Specifically, the attention level calculation unit 213 increases the attention level of the target indicated by the attention target data obtained for a certain guest according to the length of the duration of the attention of the guest to the target. The degree of attention of a plurality of exhibition objects may be calculated. For example, the attention level calculation unit 213 aggregates the number of appearances of the exhibition object in the attention target data for each exhibition object over a plurality of playback times, and calculates the attention level of the exhibition object based on the aggregation result. To do. Further, the attention level calculation unit 213 may discard the series of attention target data when the duration of attention to a certain target indicated by the series of attention target data obtained for a certain guest is less than the threshold value. .. In other words, the attention level calculation unit 213 determines the attention level based on the series of attention target data when the duration of attention to a certain target indicated by the series of attention target data obtained for a guest exceeds the threshold value. May be calculated.

The attention level calculation unit 213 may calculate the attention level for each guest / attribute. That is, the attention level calculation unit 213 reads the attention target data including a certain guest identifier / attribute identifier from the attention target data storage unit 223, and for each exhibition target, a plurality of appearance numbers of the exhibition target in the read attention target data are displayed. The data may be aggregated over the playback time, and the degree of attention of the exhibit target of the guest having the attribute value indicated by the guest / attribute identifier indicated by the guest identifier may be calculated based on the aggregation result.

Further, the attention level calculation unit 213 may calculate the attention level for each host. That is, the attention level calculation unit 213 reads the attention target data including a certain host identifier from the attention target data storage unit 223, and sets the number of appearances of the exhibition target in the read attention target data at a plurality of reproduction times for each exhibition target. It is possible to calculate the degree of attention of the exhibition target of the guest who views the video shot by the host indicated by the host identifier based on the total result.

Further, the attention level calculation unit 213 may calculate the overall attention level at the exhibition. That is, the attention level calculation unit 213 reads out all the attention target data from the attention target data storage unit 223, and totals the number of appearances of the exhibition target in the read attention target data for each exhibition target over a plurality of reproduction times. Then, based on the aggregated result, the degree of attention of the exhibition target of all the guests who virtually participate in the exhibition may be calculated.

The attention level storage unit 222 stores the attention level calculated by the attention level calculation unit 213. The attention level stored in the attention level storage unit 222 is read out by various functional units of the processor 210, for example, the attention level reporting unit 214. Note that the degree of attention may include, for example, a desired playback time, guest identifier, host identifier, and / or attribute identifier as a key, and even if one that matches the desired key can be selectively read out. Good.

The attention level reporting unit 214 reads the attention level from the attention level storage unit 222 and reports this to, for example, an attention level utilization server (not shown). Specifically, the attention level reporting unit 214 sends the attention level to the transmission unit 202, and the transmission unit 202 transmits this attention level to the attention level utilization server via the network.

Hereinafter, an operation example of the attention degree calculation server 40 will be described with reference to FIG. The operation of FIG. 9 starts from step S601.
In step S601, the receiving unit 201 waits for the viewing behavior data to be received from the guest terminal 30. When the receiving unit 201 receives the viewing behavior data, the process proceeds to step S602. On the other hand, if the attention level calculation timing arrives before the receiving unit 201 receives the viewing behavior data, the process proceeds to step S606 (step S605).

Here, the timing of calculating the degree of attention can be arbitrarily determined. For example, when the attention level is calculated in real time during the exhibition, the attention level calculation timing may be determined, for example, every 1 minute, every 10 minutes, every 1 hour, and so on. Alternatively, for example, the timing of calculating the degree of attention may be determined after the end of the half-day or one-day schedule of the exhibition, or after the end of the entire schedule of the exhibition.

In step S602, the reception data acquisition unit 211 acquires the viewing behavior data received by the receiving unit 201 and writes it in the viewing behavior data storage unit 221. Next, the attention target identification unit 212 identifies the target that the guest paid attention to at the playback time indicated by the playback time data included in the viewing behavior data based on the viewing behavior data acquired in step S602, and the attention target data. Is generated (step S603).

Note that steps S602 and S603 do not necessarily have to be performed continuously. For example, step S603 may be executed every time step S602 is executed a plurality of times or at a predetermined cycle. In this case, after step S602, the process may return to step S601 or proceed to step S605 instead of step S603.

The attention target identification unit 212 writes the attention target data generated in step S603 to the attention target data storage unit 223 (step S604), and the process proceeds to step S605. After step S604, the process may return to step S601 instead of step S605.

In step S606, the attention level calculation unit 213 calculates the attention level of a plurality of exhibition objects based on the attention target data stored in the attention target data storage unit 223 in step S604, and stores the attention level in the attention level storage unit 222. ..

Then, the attention level reporting unit 214 reports the attention level calculated in step S606 to, for example, a attention level utilization server (step S607) (not shown), and the operation of FIG. 9 ends.

Note that steps S606 and S607 do not necessarily have to be performed continuously. For example, step S607 may be executed every time step S606 is executed a plurality of times or at a predetermined cycle. In this case, after step S606, the process may return to step S601 or step S605 instead of step S607.

As described above, the attention level calculation server according to the embodiment acquires the viewing behavior data based on the input acquired at the guest terminal sharing the video data of the exhibition taken by the host, and exhibits based on the input. Identify the target of the guest's attention among the multiple exhibition targets at the meeting. Then, this attention degree calculation server calculates the attention degree of a plurality of exhibition objects so that the attention degree of the exhibition object specified as the attention object of the guest relatively increases. Therefore, since this attention level calculation server calculates the attention level of the exhibition target from the guests who virtually participate in the exhibition, the exhibition target is different from the questionnaire to the actual participants of the exhibition. You can collect data related to your interest in.

(Modification example)
In the description of the embodiment, the attention level calculation server calculates the attention level, but the guest terminal or another computer may calculate the attention level. A device that calculates the degree of attention, such as a attention degree calculation server or a guest terminal, can be called an attention degree calculation device. Further, the calculation of the degree of attention may be performed by a device different from the identification of the target of attention.

The above embodiments are merely specific examples to aid in understanding the concepts of the invention and are not intended to limit the scope of the invention. In the embodiment, various components can be added, deleted or converted without departing from the gist of the present invention.

In the above-described embodiment, some functional parts have been described, but these are only examples of implementation of each functional part. For example, it is possible that a plurality of functional parts described as being mounted on one device may be mounted on a plurality of separate devices, and conversely, it is described as being mounted on a plurality of separate devices. It is also possible that the functional unit is mounted on one device.

The various functional parts described in each of the above embodiments may be realized by using a circuit. The circuit may be a dedicated circuit that realizes a specific function, or may be a general-purpose circuit such as a processor.

At least a part of the processing of each of the above embodiments can be realized by using, for example, a processor mounted on a general-purpose computer as basic hardware. The program that realizes the above processing may be provided by storing it in a computer-readable recording medium. The program is stored on the recording medium as a file in an installable format or a file in an executable format. Examples of the recording medium include magnetic disks, optical disks (CD-ROM, CD-R, DVD, etc.), magneto-optical disks (MO, etc.), semiconductor memories, and the like. The recording medium may be any medium as long as it can store the program and can be read by a computer. Further, the program that realizes the above processing may be stored on a computer (server) connected to a network such as the Internet and downloaded to the computer (client) via the network.

10 ... Host terminal 20 ... Video distribution server 30, 30-1, 30-2 ... Guest terminal 40 ... Attention calculation server 100, 200 ... Communication I / F
101, 201 ... Reception unit 102, 202 ... Transmission unit 110, 210 ... Processor 111,211 ... Received data acquisition unit 112 ... Decoder 113 ... Video output unit 114 ... Input Data acquisition unit 115 ... Viewing behavior data generation unit 116 ... Viewing behavior reporting unit 120, 220 ... Memory 121 ... Video data storage unit 122, 221 ... Viewing behavior data storage unit 123 ... Input data storage unit 130 ・ ・ ・ Output device 140 ・ ・ ・ Input device 212 ・ ・ ・ Attention target identification unit 213 ・ ・ ・ Attention degree calculation unit 214 ・ ・ ・ Attention degree reporting unit 222 ・ ・ ・ Attention degree storage unit 223 ・・ ・ Data storage unit 301, 302 ・ ・ ・ Booth 301-1, 301-2, 301-3, 302-1 ・ ・ ・ Exhibit 401, 402 ・ ・ ・ GUI parts 501 ・ ・ ・ Host position 502・ ・ ・ Guest viewing range

Claims (10)

An acquisition unit that acquires the guest's viewing behavior data based on the input acquired at the guest's terminal that shares the video data of the exhibition taken by the host.
Based on the viewing behavior data, a specific unit that identifies the object of interest of the guest among the plurality of exhibition objects at the exhibition, and
An attention level calculation device including a calculation unit for calculating the attention level of the plurality of exhibition objects so that the attention level of the exhibition object specified as the attention target of the guest is relatively increased. The video data represents a panoramic video.
The viewing behavior data includes visual field direction data indicating the visual field direction of the guest at the first reproduction time of the video data, and reproduction time data indicating the first reproduction time.
The specific unit extracts the visual field image of the guest at the first reproduction time based on the reproduction time data and the visual field direction data from the video data, and sets the subject of the visual field image to the first reproduction time. Identify the target that the guest paid attention to,
The attention degree calculation device according to claim 1. The video data represents a panoramic video.
The viewing behavior data indicates the reproduction time data indicating the first reproduction time of the video data corresponding to the time when the guest terminal captures the image data, and the viewing direction of the guest at the first reproduction time. Including visual field data
The specific unit extracts the visual field image of the guest at the first reproduction time based on the reproduction time data and the visual field direction data from the video data, and sets the subject of the visual field image to the first reproduction time. Identify the target that the guest paid attention to,
The attention degree calculation device according to claim 1. The viewing behavior data is a reproduction time data indicating a captured image of the video data generated by the guest terminal in response to the capture operation and a first reproduction time of the video data corresponding to the capture operation when the capture operation is performed. Including and
The specific unit identifies an object that the guest has paid attention to at the first reproduction time from the subject of the captured image.
The attention degree calculation device according to claim 1. The viewing behavior data is a first GUI component selected by the guest among GUI (Graphical User Interface) components displayed superimposed on the image represented by the video data at the first reproduction time of the video data. The GUI selection data shown and the playback time data indicating the first playback time are included.
The specific unit extracts the visual field image of the guest at the first reproduction time from the video data based on at least the reproduction time data, and the first one based on the GUI selection data from the subject of the visual field image. Identify the target of interest of the guest at the playback time,
The attention degree calculation device according to claim 1. The video data represents a panoramic video.
The viewing behavior data includes visual field direction data indicating the visual field direction of the guest at the first reproduction time of the video data and reproduction time data indicating the first reproduction time.
The specific unit has a visual field range of the guest at the first reproduction time based on the reference position / orientation data indicating the position of the host at the first reproduction time and the reference angle of the video data and the visual field direction data. Is determined, and the object of interest of the guest at the first reproduction time is specified based on the determined visual field range and the map data that associates each of the plurality of exhibition objects with the position of the exhibition object.
The attention degree calculation device according to claim 1. The video data represents a panoramic video.
The viewing behavior data indicates playback time data indicating a first playback time of the video data corresponding to when the guest terminal captures the video data, and the viewing direction of the guest at the first playback time. Including visual field data
The specific unit has a visual field range of the guest at the first reproduction time based on the reference position / orientation data indicating the position of the host at the first reproduction time and the reference angle of the video data and the visual field direction data. Is determined, and the object of interest of the guest at the first reproduction time is specified based on the determined visual field range and the map data that associates each of the plurality of exhibition objects with the position of the exhibition object.
The attention degree calculation device according to claim 1. The calculation unit pays attention to the plurality of exhibition objects so that the degree of attention of the exhibition object specified as the object of interest of the guest increases according to the duration of the attention of the guest to the exhibition object. The attention degree calculation device according to any one of claims 1 to 7, wherein the degree is calculated. The computer acquires the viewing behavior data of the guest based on the input acquired at the terminal of the guest sharing the video data of the exhibition taken by the host.
Based on the viewing behavior data, the computer identifies the object of interest of the guest among the plurality of exhibition objects at the exhibition.
A method of calculating attention, which comprises calculating the attention of the plurality of exhibition objects so that the computer relatively increases the attention of the exhibition objects specified as the objects of interest of the guest. Computer,
A means for acquiring the guest's viewing behavior data based on the input acquired at the guest's terminal that shares the video data of the exhibition taken by the host.
A means for identifying an object of interest of the guest among a plurality of exhibition objects at the exhibition based on the viewing behavior data.
An attention level calculation program that functions as a means for calculating the attention levels of a plurality of exhibition objects so that the attention levels of the exhibition objects specified as the attention objects of the guests are relatively increased.