Information processing device

Abstract

To provide an information processing device which can suppress a construction cost of a system while utilizing evaluation accuracy of an advertisement using a camera.SOLUTION: An information processing device according to the present disclosure for estimating the number of people who view an out-of-home advertisement comprises: viewing information acquisition means which acquires viewing information to the advertisement based on an image obtained by photographing a person who can view the first advertisement provided at a first point; relation acquisition means which acquires a relation between the staying time of the person and an index of the number of people who view the advertisement specified on the basis of the viewing information; staying time acquisition means which acquires the staying time of the person who possesses a device that can communicate to a radio device based on a radio signal received from the radio device arranged at a second point where the second advertisement is provided; and estimation means which estimates the number of people who view the second advertisement by applying the relation to the staying time obtained for the second point. The viewing information includes the staying time of the person who can view the first advertisement and information on whether the person has viewed the first advertisement.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing apparatus for evaluating out-of-home (OOH) advertisements. [Background technology]

[0002] Conventionally, a technology is known for evaluating advertisements by attaching a camera to the advertisement to capture an image in front of it and using image processing to determine whether or not a person in the captured image has viewed the advertisement (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2000-209578 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Patent Document 1 requires a camera and an image recognition device for each advertisement to be evaluated, which poses a problem in that the cost of building the system increases as the number of advertisements to be evaluated increases.

[0005] On the other hand, a simple configuration that does not require a camera is being considered, which involves placing a communication device using a communication method such as Bluetooth (registered trademark) or Wi-Fi next to an advertisement. This aims to evaluate an advertisement by measuring the flow of people around the advertisement based on wireless communication between the communication device and a smartphone or other device carried by the person. However, simply measuring the flow of people has the problem of not being able to grasp how many people have looked at the advertisement. Furthermore, although the evaluation should change depending on the content of the advertisement, it is not possible to evaluate differences in content.

[0006] The present invention has been made in view of the above-mentioned problems, and its purpose is to realize a technology that can reduce the cost of building a system while taking advantage of the accuracy of evaluating advertisements using a camera. [Means for solving the problem]

[0007] In order to solve this problem, for example, an information processing device of the present invention has the following arrangement: An information processing device that estimates the number of people who viewed an out-of-home (OOH) advertisement, visual information acquisition means for acquiring visual information regarding the advertisement based on an image of a person who can view the first advertisement provided at the first location; a relationship acquisition means for acquiring a relationship between a person's stay time and an index of the number of people who viewed the advertisement, the relationship being specified based on the viewing information; a stay time acquisition means for acquiring a stay time of a person holding a device capable of communicating with a wireless device placed at a second location where a second advertisement is provided, based on a wireless signal received from the wireless device; an estimation means for estimating the number of people who viewed the second advertisement by applying the relationship to the stay time obtained for the second location, The viewing information includes information indicating a stay time of a person who can view the first advertisement and whether the person has viewed the first advertisement. fruit, the first advertisement and the second advertisement have the same content; The number of people who saw the ad is the percentage of people who saw the ad among those who were present at the location where the ad was served. , characterized in that [Effects of the Invention]

[0008] According to the present invention, it is possible to reduce the cost of building a system while taking advantage of the accuracy of evaluating advertisements using a camera. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an outline of an advertisement evaluation system according to an embodiment of the present invention. [Figure 2]FIG. 10 is a diagram for explaining an OOH advertisement placed together with a camera according to the present embodiment. [Figure 3] FIG. 1 is a diagram illustrating an OOH advertisement placed with a wireless device according to an embodiment of the present invention. [Figure 4] FIG. 1 is a block diagram showing an example of the hardware configuration of a server device as an example of an information processing apparatus according to an embodiment of the present invention; [Figure 5] FIG. 1 is a block diagram showing an example of the functional configuration of a server device according to an embodiment of the present invention; [Figure 6] FIG. 10 is a diagram illustrating the relationship between visual inspection probability and staying time according to the present embodiment. [Figure 7] FIG. 10 is a diagram showing an example of a data structure of visual information according to the present embodiment; [Figure 8] FIG. 10 is a diagram showing an example of a data structure of stay information according to the present embodiment; [Figure 9] FIG. 10 is a diagram showing an example of a data structure of reception history information according to the present embodiment; [Figure 10] FIG. 10 is a diagram showing an example of a data structure of relationship information according to the embodiment; [Figure 11] A flowchart showing a series of operations in the visual count estimation process according to the present embodiment. [Figure 12] FIG. 10 is a diagram showing another example of the data structure of relationship information according to the embodiment; DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the invention as claimed, and not all combinations of features described in the embodiments are necessarily essential to the invention. Two or more of the features described in the embodiments may be combined in any desired manner. Furthermore, the same reference numerals are used to designate identical or similar components, and redundant descriptions will be omitted.

[0011] <Outline of the advertising evaluation system> 1 shows an overview of an advertisement evaluation system according to this embodiment. The advertisement evaluation system includes, for example, a server device 100, a camera 103, an edge device 104, and a wireless device 106.

[0012] Advertisement 101 is an OOH advertisement placed at a first location together with (for example, adjacent to) camera 103, and includes still image advertisements and video advertisements displayed on a monitor or the like. Advertisement 101 also includes paper or sheet-like advertisements placed on a wall or the like that do not rely on display means such as a monitor. Advertisement 102 is an OOH advertisement placed at a second location together with wireless device 106. Like advertisement 101, advertisement 102 also includes still image advertisements and video advertisements displayed on a monitor or the like, or advertisements placed on a wall or the like that do not rely on display means.

[0013] 2, the camera 103 is a camera that captures an image in front of the advertisement 101, and is equipped with, for example, a wide-angle lens. The camera 103 can capture, as successive frames, an image of a person 105 who can see the advertisement 101 passing in front of the advertisement 101 or stopping in front of the advertisement 101, and output the captured images sequentially. The camera 103 may be integrated with the edge device 104.

[0014] The edge device 104 may be, for example, a smartphone, a tablet computer, a personal computer, or the like. The edge device applies image recognition processing to the image captured by the camera 103 to generate visual observation information (described later). The edge device recognizes and tracks the people 105 in the image using, for example, a pre-trained deep neural network (DNN) to determine the time (e.g., seconds) that each person 105 in the image spent at a first location. In this embodiment, the edge device 104 determines, for example, the time that each person in the image spent in the image (or in a predetermined area including the center of the image) as the stay time of each person at the first location. The edge device 104 also detects the gaze direction of the person in the image to determine whether the person looked at the position of the advertisement 101. Based on the recognition result, the edge device 104 can determine whether the person viewed the advertisement 101. The recognition and tracking of people in the image and the determination of whether they viewed the advertisement can be achieved using known technologies. The recognition and tracking of people in images and the determination of whether they have viewed an advertisement can be processed using one or more DNNs. Furthermore, the recognition and tracking of people in images and the determination of whether they have viewed an advertisement can also be performed using other machine learning techniques or other recognition modules.

[0015] The edge device 104 recognizes and tracks people in the image and determines whether they have looked at the advertisement 101, thereby generating visual observation information including information indicating the duration of stay of each person at the first location and whether each person has looked at the advertisement 101. The edge device 104 transmits the generated visual observation information to the server device 100.

[0016] As shown in FIG. 3 , the wireless device 106 is placed together with the advertisement 102. The wireless device 106 transmits a probe request signal, for example, via Bluetooth (registered trademark) or WiFi, to a mobile device 108 carried by a person 107 passing by the vicinity of the advertisement 102. When the mobile device 108 receives an advertisement packet including, for example, an identifier (e.g., a UUID) of the wireless device 106 from the wireless device 106, the mobile device 108 transmits reception information including the identifier to a predetermined device. For simplicity, the example shown in FIG. 1 illustrates the mobile device 108 transmitting the reception information to the wireless device 106. However, the predetermined device receiving the reception information may be a location information server (not shown). The location information server or the wireless device 106 generates reception history information based on the information from the mobile device 108 and transmits it to the server device 100. The reception history information will be described later.

[0017] Referring again to FIG. 1 , the server device 100 is one or more servers as an example of an information processing device, and executes the below-described viewing count estimation process. For example, the server device 100 communicates with the edge device 104 and the wireless device 106 via a network to acquire viewing information and reception history information. After acquiring the viewing information, the server device 100 acquires a relationship between a person's stay time and an index of the number of people who viewed the advertisement, which is obtained from the viewing information. Note that the "relationship between a person's stay time and an index of the number of people who viewed the advertisement" is also simply referred to as the "relationship between stay time and viewing probability." This relationship may be acquired by applying machine learning to the index of a person's stay time and the number of people who viewed the advertisement, or the relationship determined based on the viewing information may be input from the administrator of the server device 100. The index of the number of people who viewed the advertisement indicates the proportion of people who viewed the advertisement among people who stayed at the location where the advertisement was provided. The server device 100 can calculate the number of people who viewed the advertisement 101 from the viewing information.

[0018] Furthermore, the server device 100 can calculate the stay time of a person (i.e., at the second location) who holds a device that can communicate with the wireless device 106, from the reception history information. The stay time calculated from the reception history information can be, for example, the time that a device held by the person stays in a range where a signal from the wireless device 106 is received with a predetermined radio wave intensity or higher. Furthermore, the server device 100 applies the relationship between stay time and visual probability obtained for the first location to the stay time of the person obtained for the second location. This makes it possible to estimate the number of people who viewed the advertisement at the second location. Note that in this embodiment, it is assumed that the administrator has previously set which location's relationship between stay time and visual probability to apply in order to estimate the number of people who viewed the advertisement at the second location.

[0019] <Server device configuration> Next, an example of the hardware configuration of the server device 100 will be described with reference to Fig. 4. The server device 100 includes a memory 401, a processor 402, a communication interface 403, an input interface 404, and a storage 405. These elements are each connected to a bus and communicate with each other via the bus.

[0020] The memory 401 is a volatile storage medium for temporarily storing data and computer programs. The storage 405 is a non-volatile storage medium for permanently storing data and computer programs. The storage 405 also stores data described below in the functional configuration example of the server device 100.

[0021] The processor 402 includes one or more processors such as a CPU, and implements various functions of the server device 100 by loading computer programs stored in the storage 405 into the memory 401 and executing them. The communication interface 403 is an interface for transmitting and receiving data to and from the outside of the server device 100. The communication interface 403 is connected to a network, and exchanges data with the edge device 104, the wireless device 106, or a location information server (not shown) via the network. The input interface 404 is a device for receiving input from an administrator of the server device 100, but may be omitted.

[0022] Next, an example of the functional configuration of the server device 100 will be described with reference to Figure 5. Each block shown here can be realized in hardware by elements, circuits, and mechanical devices such as a computer CPU, and in software by a computer program, etc., but here, functional blocks realized by the cooperation of these elements are depicted. Therefore, these functional blocks can be realized in various forms by combining hardware and software. Furthermore, each piece of information shown here is temporarily or permanently stored in at least one of the memory and storage described below.

[0023] The visual inspection information acquisition unit 501 acquires the visual inspection information generated in the edge device 104 and stores the visual inspection information in a database 510. The visual inspection information acquisition unit 501 may acquire the visual inspection information in real time or at predetermined time intervals.

[0024] The relationship acquisition unit 502 acquires the relationship between the stay time of a person and the index of the number of people who viewed the advertisement (relationship between stay time and viewing probability), which is obtained from the viewing information. Here, the index of the number of people who viewed the advertisement indicates, for example, the proportion of people who viewed the advertisement among people who stayed at the location (visual probability).

[0025] FIG. 6 shows a schematic example of the relationship between stay time and visual probability. In FIG. 6, the horizontal axis indicates a person's stay time, and the vertical axis indicates the visual probability. That is, 602 indicates an index of the number of people who viewed an advertisement for each person's stay time (visual probability). Relationship 601 is a curve specified by, for example, a function, that is determined to fit the visual probability for each stay time. By using relationship 601, it is possible to calculate the visual probability for each stay time, even if samples for some stay times have not been obtained.

[0026] The server device 104 may acquire the relationship 601 by applying machine learning to the indicators of the person's stay time and the number of people who viewed the advertisement. Alternatively, the relationship 601 may be acquired as an input from the administrator of the server device 100, where the relationship 601 is determined separately (by the administrator using a statistical method) based on the viewing information.

[0027] The reception history acquisition unit 503 acquires reception history information from the wireless device 106 or a location information server (not shown) and stores it in the database 510. The staying time acquisition unit 504 calculates the staying time of a person who holds a device that can communicate with the wireless device 106, from the reception history information. As described above, the staying time calculated from the reception history information may be, for example, the time that a device held by a person stays within a range that receives a signal from the wireless device 106 with a predetermined radio wave intensity or higher.

[0028] The viewing number estimation unit 505 estimates the number of people who viewed the advertisement at the second location by applying the relationship between the stay time (obtained for the first location) and the viewing probability to the stay time of the person (at the second location) calculated by the stay time acquisition unit 504. For example, the viewing number estimation unit 505 multiplies the number of people for each stay time at the second location by the viewing probability for each stay time at the first location to calculate the number of people who viewed the advertisement for each stay time, and adds up the obtained number of people who viewed the advertisement.

[0029] The view count providing unit 506 provides information on the number of people who viewed an advertisement to an external device. For example, when the view count providing unit 506 receives a request from a client device (not shown) to acquire the number of people who viewed a specific advertisement provided at points A and B, the view count providing unit 506 provides, for example, a list of the number of people who viewed the specific advertisement for each point. In this way, the business that placed the advertisement or its advertising agent can evaluate the advertisement for each point where the advertisement was provided.

[0030] The database 510 is, for example, a database, and includes, for example, a visual inspection information table 511, a stay time information table 512, a reception history information table 513, and a relationship information table 514.

[0031] The visual observation information table 511 will be described with reference to FIG. 7. FIG. 7 shows an example of the data structure of the visual observation information table 511. The visual observation information includes a user ID, a location, an acquisition time, information indicating whether or not the person has visually observed the advertisement, and a stay time. The user ID is an identifier for identifying a person. In this example, the user ID is assigned simply to distinguish between people. The location is information specifying a location (equipped with a camera 103) where an advertisement was provided. The acquisition time is, for example, the time when the visual observation information of person A1 was generated, but may be another time. For example, if the advertisement has been visually observed, the acquisition time may be the first time the visual observation occurred, or may be the time when the person 105 started staying at location A. The presence or absence of visual observation indicates whether or not the person 105 has visually observed the advertisement through the recognition process of the edge device 104. The stay time indicates the time that the person 105 (recognized by the edge device) stayed at location A. As described above, the stay time at location A is the time that each person in the image stayed in the image (or in a predetermined area in the image).

[0032] The staying time information table 512 will be described with reference to FIG. 8. FIG. 8 shows an example of the data structure of the staying time information table 512. The staying time information table 512 includes a user ID, a location, an acquisition time, a MAC address, and a staying time. The user ID is an identifier for identifying a person. In this example, the user ID is assigned simply to distinguish between people. The location is information specifying a location (including a wireless device) where an OOH advertisement was provided. The acquisition time is, for example, the time when the staying time information of person B1 was generated, but may be another time. For example, it may be the time when the person started staying at location B. The MAC address is the MAC address of the mobile device 108 held by the person 107. The staying time indicates the time the person 107 stayed at location B. As described above, the staying time is calculated from the reception history information, and is, for example, the time when the device held by the person stayed within a range where it can communicate with the wireless device 106 at a predetermined radio wave strength or higher.

[0033] The reception history information table 513 will be described with reference to Fig. 9. Fig. 9 shows an example of the data structure of the reception history information table 513. The reception history information table 513 includes acquisition information, a UUID, a MAC address, and radio wave strength. The acquisition time is, for example, the time when the acquisition time for a specific mobile device 108 was generated. The UUID is an identifier that identifies the wireless device 106 that transmits the advertising signal, and the MAC address indicates the MAC address of the mobile device 108. The radio wave strength indicates, for example, the radio wave strength when the mobile device 108 received the advertisement from the wireless device 106.

[0034] The relationship information table 514 will be described with reference to FIG. 10. FIG. 10 shows an example of the data structure of the relationship information table 514. The relationship information table 514 includes a relationship identifier, a target advertisement ID, a location, and a relationship between stay time and visual probability. The relationship between stay time and visual probability acquired by the relationship acquisition unit 502 is acquired for each target advertisement because the number of people who view an advertisement may vary depending on the content of the advertisement. The relationship identifier is an identifier that identifies the relationship between stay time and visual probability acquired by the relationship acquisition unit 502. The target advertisement ID is an identifier of the advertisement corresponding to the relationship between stay time and visual probability. The location indicates the location where the advertisement was provided. The relationship between stay time and visual probability is information that specifies the relationship between stay time and visual probability, and is, for example, information such as a coefficient that specifies a function (i.e., relationship 601). When the relationship between stay time and visual probability is specified by a DNN, the relationship identifier may be a hyperparameter or a learned weight coefficient of the DNN.

[0035] <Edge device configuration> The edge device 104 may have a hardware configuration similar to that shown in Fig. 4. For example, the edge device 104 executes the above-described recognition and tracking of people in images and determination of whether or not an advertisement has been viewed by executing a computer program stored in storage using one or more processors.

[0036] <Wireless device configuration> The wireless device 106 may have a hardware configuration similar to that shown in Fig. 4. For example, the wireless device 106 transmits a probe request signal to nearby mobile devices 108 via a communication interface by executing a computer program stored in storage by one or more processors.

[0037] Next, a series of operations of the visual count estimation process executed in the server device 100 will be described with reference to Fig. 11. Note that this process is realized by the processor 402 executing a computer program recorded in the storage 405. In the following description, for ease of explanation, the processing entity of each step will be collectively described as the server device 100, but each block within the server device 100 executes the corresponding process depending on the process content.

[0038] In S1101, the server device 100 acquires visual inspection information from the edge device 104 at the first location via the network. The server device 100 may store the acquired visual inspection information in the database 510.

[0039] In S1102, the server device 100 acquires the relationship between the stay time and the index of the number of viewers (the relationship between the stay time and the viewing probability) based on the acquired visual observation information. The server device 100 counts the number of users (number of people) for the stay time at each predetermined time interval (for example, every second) using the visual observation information data, and calculates the number of people for each stay time at the first location. Furthermore, the server device 100 counts the number of people who viewed the advertisement among the number of people for each stay time. This makes it possible to obtain an index of the number of people who viewed the advertisement for each stay time (the proportion of people who viewed the advertisement among people who stayed).

[0040] Next, the server device 100 may acquire the relationship between the stay time and the visual probability by applying machine learning to an index of the person's stay time and the number of people who viewed the advertisement. Alternatively, the relationship between the stay time and the visual probability determined based on the visual information may be acquired as an input from an administrator of the server device 100. The relationship between the stay time and the visual probability is, for example, a function of the stay time and the number of people who viewed the advertisement, and is represented by the curve 601 shown in FIG. 6. The server device 100 may register the acquired relationship in the database 510.

[0041] Furthermore, the server device 100 may calculate the number of people who viewed the first advertisement provided at the first location. For example, the server device 100 can calculate the number of people who viewed the first advertisement by summing up the number of people who viewed the advertisement over all stay times.

[0042] In S1103, the server device 100 acquires reception history information from a wireless device (or a location information server, not shown) at the second location. In S1104, the server device 100 calculates the person's stay time at the second location based on the reception history information. Among the data in the reception history information, the MAC address corresponds to the device held by the person, and the UUID corresponds to the location. Therefore, the server device 100 can calculate the person's stay time by tracking the acquisition time of data in which the UUID and MAC address have specific values ​​and the radio wave intensity is higher than a predetermined threshold.

[0043] In S1105, the server device 100 estimates the number of people who viewed the advertisement at the second location based on the person's stay time and the relationship between the stay time and the viewing probability. The relationship between the stay time and the viewing probability is a function that takes the stay time and the number of people as input, so the number of people who viewed the advertisement at the second location can be estimated by inputting the stay time and the number of people measured at the second location into the function. Note that, when a trained machine learning model is used, the number of people who viewed the advertisement at the second location can be estimated by inputting the stay time and the number of people into the machine learning model.

[0044] In S1106, the server device 100 provides the number of viewers of the advertisement estimated in S1105 to an external device (not shown). The external device (not shown) is a client device used by an advertiser, an advertising agency, or the like.

[0045] The server device 100 may provide not only information about the location where the wireless device 106 is installed, but also the number of viewers at the location where the camera is installed. In the description of this process, an example is given where the wireless device 106 is installed at one location, but if the wireless device 106 is installed at multiple locations, the number of viewers calculated for each location may be provided as a list to the external device. In this way, it is possible to provide a combined number of viewers of the advertisement at multiple locations. After providing the number of viewers of the advertisement to the external device, the server device 100 ends this series of processes.

[0046] As described above, in this embodiment, visual observation information for the advertisement is obtained based on an image captured of a person who can view the advertisement 101. Also, the relationship between stay time and visual observation probability, which is determined based on the visual observation information, is obtained. Furthermore, the stay time of a person holding a device 108 that can communicate with the wireless device 106, is obtained based on a wireless signal received from the wireless device 106. Then, the relationship between the stay time and the visual observation probability is applied to the stay time obtained for the advertisement 102, thereby estimating the number of people who viewed the advertisement 102. In this way, it is possible to reduce the system construction cost while taking advantage of the accuracy of advertisement evaluation using a camera.

[0047] In the above-described embodiment, the advertisement 101 and the advertisement 102 do not change over time. However, the above-described embodiment can also be applied to a case where the advertisement 101 is switched to a different advertisement over time. For example, if advertisement A is provided from 6:00 to 15:00 and advertisement B is provided from 15:00 to 24:00, the server device 100 acquires the relationship between the stay time and the viewing probability for each advertisement. For example, the relationship information table 514 described above in FIG. 10 may be modified as shown in FIG. 12. FIG. 12 shows another example of the data structure of the relationship information table 514. The relationship information table 514 includes a relationship identifier, a target advertisement ID, a location, a time, and a relationship between the stay time and the viewing probability. Since different advertisement contents may result in different numbers of people viewing the advertisement, the relationship between the stay time and the viewing probability is acquired for each target advertisement. The relationship identifier is an identifier that identifies the relationship between the stay time and the viewing probability acquired by the relationship acquisition unit 502. The information other than "time" is the same as that shown in FIG. 10. The time indicates when the advertisement is served at point A.

[0048] The server device 104 estimates the number of people who viewed advertisement A by applying the relationship between the stay time and the viewing probability for advertisement A to the stay information obtained at the second location (where advertisement A is provided). Meanwhile, the server device 104 estimates the number of people who viewed advertisement B by using the relationship between the stay time and the viewing probability for advertisement B to the stay information obtained at the third location (where advertisement B is provided). In this way, it is possible to obtain the relationship between the stay time and the viewing probability for multiple advertisements at one location, and it becomes possible to estimate the number of people who viewed each of different advertisements at multiple locations. In other words, the number of advertisements for which cameras need to be installed can be reduced, further reducing the system construction costs.

[0049] In the above-described embodiment, since the calculation method of the dwell time obtained using the camera 103 differs from the calculation method of the dwell time obtained using the wireless device 106, similar distributions of dwell times may not be obtained for similar human flows. For this reason, a correction method determined in advance through experiments or the like may be used to correct the distribution of dwell times obtained using the wireless device 106 for the same human flow so that it approaches the distribution of dwell times obtained using the camera 103. Alternatively, conversely, the distribution of dwell times obtained using the camera 103 may be corrected.

[0050] In the above embodiment, an example was described in which an administrator preliminarily sets the relationship between the stay time and the visual probability obtained at the first location to be applied to estimate the number of people who viewed the advertisement at the second location. However, if the advertisement 101 with the camera 103 placed thereon is provided at multiple locations, the server device 100 may determine which location's relationship between the stay time and the visual probability to apply. For example, if the stay times are obtained using the camera 103 at the first and third locations, the distribution of these stay times (regardless of whether they were visually observed) that is closest to the distribution of stay times at the second location using a wireless device may be used. Alternatively, the distribution of stay times obtained using the camera 103 may be corrected using the above-mentioned correction, and then the distribution that is closest to the distribution of stay times at the second location may be used.

[0051] Furthermore, in the above-described embodiment, an example has been described in which the wireless device 106 is not placed at the first location where the camera 103 is placed. However, the wireless device 106 may also be placed at the first location to measure the stay time at the first location. If the advertisement 101 with the camera 103 placed therein is provided at a plurality of locations, the relationship between the stay time and the visual probability at the location closest to the distribution of stay time at the second location, among the distributions of stay time obtained using the wireless device 106, may be used.

[0052] The invention is not limited to the above-described embodiment, and various modifications and variations are possible within the scope of the gist of the invention. [Explanation of symbols]

[0053] 100...server device, 101, 102...OH advertisement, 104...edge device, 106...wireless device

Claims

1. An information processing device that estimates the number of people who viewed an out-of-home (OOH) advertisement, visual information acquisition means for acquiring visual information regarding the advertisement based on an image of a person who can view the first advertisement provided at the first location; a relationship acquisition means for acquiring a relationship between a person's stay time and an index of the number of people who viewed the advertisement, the relationship being specified based on the viewing information; a stay time acquisition means for acquiring a stay time of a person holding a device capable of communicating with a wireless device disposed at a second location where a second advertisement is provided, based on a wireless signal received from the wireless device; an estimation means for estimating the number of people who viewed the second advertisement by applying the relationship to the stay time obtained for the second location, the viewing information includes information indicating a stay time of a person who can view the first advertisement and whether the person has viewed the first advertisement; the first advertisement and the second advertisement have the same content, An information processing device characterized in that the index of the number of people who viewed an advertisement represents the ratio of people who viewed the advertisement to people who stayed at a location where the advertisement was provided.

2. The staying time acquisition means a reception history acquisition means for acquiring reception history information based on the wireless signal received from the wireless device; 2. The information processing apparatus according to claim 1, further comprising: a calculation unit that calculates the stay time of a person who has a device capable of communicating with the wireless device from the reception history information.

3. The information processing apparatus according to claim 2 , wherein the reception history information includes radio wave intensity received from the wireless device by a communication device held by a person who can view the second advertisement.

4. The information processing device according to claim 1, characterized in that the relationship acquisition means acquires, for each advertisement, the relationship between a person's stay time and an indicator of the number of people who viewed the advertisement when the advertisements provided at the first location are switched to different advertisements over time.

Images