JP2023025990A - Information processing apparatus, information processing method, and program - Google Patents

Abstract

To allow for reducing user's time and effort and the load of analysis processing.SOLUTION: An information processing apparatus determines a size of an area where estimation is possible by prescribed estimation processing, on the basis of a size and position of an area designated by a user on a setting screen where an image acquired by image capture is displayed. Then, the information processing apparatus displays an index representing the size of the area where estimation is possible by the estimation processing, on the setting screen for the user to designate the size and position of the area.SELECTED DRAWING: Figure 3

Description

The present invention relates to information processing technology for analyzing captured images.

Patent Literature 1 discloses a technique for detecting specific objects such as people from an image and estimating the number of such objects. Further, in Patent Document 1, for a region where the size and position of the human body cannot be obtained from the image, based on the average size of the human body instructed by the user, the average size of the human body at an arbitrary position on the image is Techniques for estimating magnitude by interpolation are disclosed.
Further, Patent Literature 2 discloses a technique of presenting information necessary for appropriately performing setting work and parameter correction work for obtaining desired processing results to a user.

Japanese Patent Application Laid-Open No. 2018-22340 JP 2016-143157 A

When image analysis is performed to obtain the result of estimation processing such as detection of a specific object from an image, whether or not the estimation result expected by the user can be obtained cannot be determined until the result of the actual image analysis is confirmed. I don't know. Then, if the estimation result is not the result expected by the user, the user needs to repeat the work of setting or adjusting the values used in the estimation process and re-executing the image analysis, which is very time-consuming. In addition, the processing load of image analysis also increases.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to make it possible to reduce the user's trouble and the load of analysis processing.

The information processing apparatus according to the present invention is capable of estimating the size of an area that can be estimated by a predetermined estimation process based on the size and position of an area specified by a user on a setting screen on which an acquired image is displayed. and an index representing the size of the area that can be estimated by the estimation process determined by the determining means on the setting screen for the user to specify the size and position of the area. and display means for displaying.

According to the present invention, it is possible to reduce user's trouble and load of analysis processing.

It is a figure which shows the whole structural example of an information processing system. It is a figure which shows the hardware structural example of an information processing apparatus. It is a figure which shows an example of the functional structure of an information processing apparatus. It is a flowchart which shows the flow of information processing. FIG. 10 is a diagram showing an example of a screen for accepting input of face size and face position; FIG. 10 is a diagram showing an example of a screen that superimposes an estimable face size; FIG. 10 is a diagram used for explaining the positions where the estimable face sizes are superimposed and displayed; FIG. 10 is a diagram showing an example of an icon that superimposes an estimable face size;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments according to the present invention will be described in detail with reference to the drawings. The following embodiments do not limit the present invention, and not all combinations of features described in the embodiments are essential to the solution of the present invention. The configuration of the embodiment can be appropriately modified or changed according to the specifications of the device to which the present invention is applied and various conditions (use conditions, use environment, etc.).

Before describing the details of the configuration and processing of the information processing apparatus of this embodiment, an application example of this embodiment will be briefly described.
2. Description of the Related Art There is an image analysis technique that makes it possible to detect predetermined objects from images acquired by imaging devices such as surveillance cameras and network cameras, and to count the number of such objects. Here, assuming a person (human body) as an example of the target object, the image analysis technology can be used to detect congestion at an event venue, etc., and furthermore, it can be used to relieve congestion and provide evacuation guidance. It becomes possible. Human body detection methods such as detection by background subtraction and pattern extraction using a human body model are known as methods for detecting people in an image. It is also possible to count the number of people.

There is also image analysis processing that directly estimates the number of people in a given area of an image using a learning model obtained by machine learning. However, in the case of image analysis processing that directly estimates the number of people using a learning model, the accuracy of estimation may change depending on how the learning model is set for the image. Generally, the size and position of a person in an image are obtained by the above-described human body detection method, and a learning model is applied according to the obtained human body size. Even in the case of using the method of applying the learning model according to the size of the human body, it is necessary for the user to set how the learning model is applied to the image. That is, the user needs to set values (parameters) for determining how to set the learning model for the image.

In this case, the user needs to check the result obtained by the image analysis processing and repeat the operation of changing the parameters to adjust the parameters appropriately. In other words, when the user sets the application parameters of the learning model, it is impossible to know whether the expected analysis results will be obtained until the actual analysis processing is performed. must be repeated. This imposes a very heavy burden on the user. As in the method disclosed in Patent Document 2, it is conceivable to present to the user information necessary for the user to appropriately perform setting work and parameter correction work for obtaining desired processing results. . In the case of this method, the user's work until the user confirms the result of the analysis processing and corrects the parameters can be reduced to some extent, but the work of executing the image analysis processing does not change. Therefore, if the image analysis process takes a long time, the user has to wait until the analysis process is completed, which impairs usability. Moreover, since it is necessary to execute image analysis processing, the processing load is not reduced.

Therefore, the information processing apparatus according to the present embodiment has the configuration described below and performs the processing described later, thereby making it possible to reduce the user's trouble and the load of the image analysis processing.
<General configuration of the system>
FIG. 1 is a diagram showing a schematic configuration example of an information processing system 101 as an application example of an information processing apparatus according to this embodiment. It is assumed that the information processing system 101 is constructed for various places such as outdoors such as theaters, stadiums, shopping malls and parks, and indoors such as meeting rooms and waiting rooms.

An information processing system 101 in FIG. 1 includes a client PC (personal computer) 102, an analysis server 103, a recording server 104, a plurality of network cameras 105, and the like. These client PC 102 to network camera 105 are connected to a network and are capable of communicating with each other. Although the client PC 102, the analysis server 103, and the recording server 104 are configured separately in FIG. 1, each function of the client PC 102 to the recording server 104 may be integrated into one PC or server. .

The client PC 102 is a personal computer or mobile terminal that can receive input from a user, issue instructions to the analysis server 103, recording server 104, and network camera 105, and display the results.

The analysis server 103 is an information processing server that receives instructions from the client PC 102 and performs image analysis processing based on the instructions. In this embodiment, the image analysis processing includes discrimination (detection) of an object such as a person appearing in an image acquired from the recording server 104 or the network camera 105, analysis of the movement of the object (person's behavior), Furthermore, it is assumed that the processing is to estimate the number of objects (number of people).

The recording server 104 is a recording server that performs recording processing based on instructions from the client PC 102 or the analysis server 103 . The recording process is a process of making an image acquisition request to a plurality of network cameras 105 connected to the same network as the recording server 104, and storing the image sent from the network camera 105 in response to the request. Point. Recording processing also includes processing for storing meta information obtained from the network camera 105 . Meta information is information including at least one of time information including time, location information of the network camera 105, setting information of the network camera 105, information of sensors attached to the network camera 105, and the like.

The network camera 105 is an imaging device, such as a monitoring camera, which is installed on the ceiling, wall, utility pole, or the like of a building, acquires an image of the surroundings, and can transmit the image to the recording server 104 via the network. Note that the image from the network camera 105 may be sent directly to the analysis server 103 or client PC 102 . The network camera 105 receives an instruction from the client PC 102, acquires images and sounds, and controls rotation of the camera. Various sensors such as a motion sensor, a temperature sensor, an audio sensor, and an infrared sensor may be connected to the network camera 105 in order to acquire the aforementioned meta information.

<Hardware configuration>
FIG. 2 is a block diagram showing a hardware configuration example of the information processing apparatus according to this embodiment. The client PC 102, analysis server 103, and recording server 104 of the information processing system 101 shown in FIG. It has a hardware configuration like The configuration of FIG. 2 is also applied to a single PC or server in which the functions of the client PC 102 to recording server 104 are integrated. In this embodiment, the information processing apparatus is implemented by a personal computer (PC), but may be implemented by an embedded system, a tablet terminal, or the like.

In FIG. 2, a CPU 201 is a central processing unit, and based on a computer program, cooperates with other components, controls the operation of the entire information processing apparatus, and executes various information processing.
A ROM 202 is a read-only memory, and stores basic programs and data used for basic processing. A RAM 203 is a writable memory and functions as a work area for the CPU 201 and the like. The external storage device 204 can access a recording medium and load computer programs and data stored in a medium (recording medium) such as a USB memory into the information processing apparatus. The storage unit 205 is a device that functions as a large-capacity memory such as an SSD (Solid State Drive). The storage unit 205 stores various programs including computer programs for information processing according to the present embodiment, and various data including image data.

An operation unit 206 is a device that receives an instruction or command input from a user, and corresponds to a keyboard, pointing device, touch panel, or the like.
A display 207 is a display device that displays a command input from the operation unit 206, a response output from the information processing apparatus in response to the command, and the like.
An interface (I/F) unit 208 is a device that relays data exchange with an external device.
A system bus 209 is a data bus that controls the flow of data within the information processing apparatus.
It should be noted that the functions of the respective units described above can also be configured as substitutes for hardware devices by software that implements functions equivalent to those of the respective units.

<Functional Configuration of Information Processing Device>
FIG. 3 is a block diagram showing the functional configuration of the information processing apparatus 301 of this embodiment realized by the client PC 102, analysis server 103, and recording server 104. As shown in FIG. In the case of this embodiment, the configuration of each functional unit shown in FIG. 3 is assumed to be implemented by executing a program. may be implemented. In the following description, an example of image analysis processing for detecting persons as objects from an image and estimating the number of such persons will be described.

The image acquisition unit 302 is a functional unit implemented in the recording server 104 . The image acquisition unit 302 sends an image acquisition request to the network camera 105 and acquires an image sent from the network camera 105 in response to the request. In this embodiment, the image captured by the network camera 105 is stored in the recording server 104, but it may be directly sent to the analysis server 103 that performs subsequent processing without being stored. A function of 302 is included in the analysis server 103 .

A display unit 306 is a functional unit implemented in the client PC 102 . The display unit 306 causes the display 207 of the client PC 102 to display a setting screen including an image captured by the network camera 105 and acquired by the image acquisition unit 302 . Details of the setting screen and processing in the display unit 306 will be described later.

The estimation unit 304 is a functional unit implemented in the analysis server 103 . The estimation unit 304 performs predetermined estimation processing on the image received from the image acquisition unit 302 and outputs the estimation result. In the present embodiment, the predetermined estimation process is a process of detecting a person as a predetermined object and estimating the number of persons (number of people) from an image acquired by imaging using a learning model. The estimating unit 304 of the present embodiment applies a regressor using a machine learning method to a region to which a learning model determined by a region determining unit 303 (to be described later) is applied to the image received from the image acquiring unit 302 to calculate the number of people. Perform the process of estimating Here, as an example, a method using a regressor whose input is an image of a fixed size and whose output is the number of people appearing in the image will be described. In this case, a large number of images in which the number of people is known in advance are used as learning data, and a known machine learning method such as deep learning is used to learn the regressor. Then, the estimation unit 304 resizes the area to which the learning model determined by the area determination unit 303 (to be described later) is applied to a fixed size, uses it as an input to the regressor, and estimates the number of people in each area. Note that the number of people to be estimated is not necessarily an integer, and may be a real number.

The integration unit 305 is a functional unit implemented in the analysis server 103 . The integration unit 305 acquires the total number of people in the image by integrating the results estimated by the estimation unit 304 for each area to which the learning model determined by the area determination unit 303 (to be described later) is applied.

The area determining unit 303 is a functional unit implemented in the analysis server 103. FIG. The area determination unit 303 performs the calculation performed by the estimation unit 304 based on the image from the image acquisition unit 302 and the size and position of the area specified by the user on the setting screen displayed on the display unit 306 . Determine the size of the region that can be estimated by a predetermined estimation process. In the case of this embodiment, the area specified by the user on the setting screen is an area including at least the face of the person, and the user specifies the size of the face (face size) and its position (face position). specify. Therefore, based on the image from the image acquisition unit 302 and the face size and face position specified by the user on the setting screen, the region determination unit 303 uses the region to which the estimation unit 304 applies the learning model, that is, the regressor. Determines the size of the area to which the number-of-people estimation process is applied. At this time, the precision of the regressor is improved by setting the ratio between the area size and the face size to be substantially constant.

The area determining unit 303 also calculates an assumed face size and an assumed face position, which will be described later. Information on the assumed face size and assumed face position is sent to the display unit 306 .
A display unit 306 is a functional unit implemented in the client PC 102 . The display unit 306 receives the information on the assumed face size and assumed face position from the area determination unit 303, and displays the indices representing the assumed face size and assumed face position superimposed on the image on the setting screen. Further, when the integration unit 305 acquires the result of counting the number of people, the display unit 306 receives and displays the result of counting the number of people.

<Flowchart of information processing>
FIG. 4 is a flowchart showing the flow of processing in the information processing system 101. As shown in FIG. Here, the image acquisition unit 302 acquires an image, the region determination unit 303 receives an instruction from the user, determines the region to which the learning model is applied, and the estimation unit 304 performs population estimation processing for each region, and integrates the images. A flow of processing in which the unit 305 acquires the result of estimating the number of people in the image will be described. Also, reference is made to FIGS. 5-8 for explanation as appropriate. In the following description of the flowcharts, processing steps are represented by symbol S.

In S401, the image acquisition unit 302 receives an image to be used for population estimation. In this embodiment, the image acquired by the image acquisition unit 302 is an image stored in the recording server 104, but may be an image directly sent from the network camera 105 or an image specified by the user. may

Next, in step S402, the area determination unit 303 selects the face size and face position information input by the user via the operation unit 206 on the setting screen displayed by the display unit 306 on the display 207 of the client PC 102. accept. Note that the face size may be any scale as long as it can measure the size of the face. Coordinates in the image are used for the face position.

FIG. 5 is a diagram showing an example of a setting screen that is displayed on the display 207 of the client PC 102 and receives input of face size and face position from the user via the operation unit 206. As shown in FIG.
A setting screen 501 is an example of a setting screen that accepts input of a face size and a face position by the user. This setting screen 501 can accept a plurality of face size and face position inputs required for setting a learning model application region (learning model application region). In addition, in the screen example of FIG. 5, an example in which two of the face size and the face position are accepted is given, but three or more may be accepted. Also, as long as the learning model application area can be set by the information, what is specified by the user does not have to be the face size and the face position. Anything can be used as long as it can calculate the learning model application area from .

Cursors 502 and 503 are cursors for specifying the position and size displayed when specifying the face position and face size. By operating the operation unit 206 of the client PC 102, the user can place the cursor on the person in the image on the setting screen 501 and specify the face position and face size. Also, the user can move the cursor by drag-and-drop or the like via the operation unit 206 of the client PC 102, and can also enlarge or reduce the size by dragging the end of the cursor. The method of operating the cursor is not limited to these examples, and any method may be used as long as it can be performed by mouse operation or touch operation. Note that FIG. 5 shows an example in which the cursor 502 is aligned with a person appearing on the near side of the screen, and the cursor 503 is aligned with a person appearing on the far side of the screen.

Returning to the flowchart of FIG. 4, upon receiving the face size and face position obtained in S402, the region determination unit 303 determines (sets) a learning model application region based on the face size and face position in S403. do. Here, the learning model application area is appropriately called a patch. In the process of setting the learning model application area (patch), the plane on which the person is standing is estimated based on the face size and face position, and the face size of the person standing on that plane is the appropriate size for the learning model. Find the patch size and position to divide the image so that Note that an area estimated to have no flat surface may be excluded from the population estimation process using the learning model, or the minimum patch size may be set. Examples of flat surfaces include the ground and floor surfaces, and examples of regions that are presumed to have no flat surfaces include wall surfaces and places where people cannot exist. Furthermore, the case of setting the minimum patch size is an example assuming a case where, for example, the distance from the photographing location by the network camera is long, and there is a high possibility that even if an object such as a person is photographed, it will be photographed extremely small.

Next, in S404, the area determining unit 303 calculates an assumed face size and an assumed face position from the size of the learning model application area obtained in S403. Here, the assumed face refers to a face that is assumed to be obtained as a result of performing the population estimation process. Since the assumed face is a face obtained as a result of the estimation process, the size of the assumed face has a range of values. The assumed face size can be obtained by obtaining the range of face sizes that can be taken by the regressor for estimating the number of people, and converting the obtained value into the ratio of the patch image. These assumed face size and assumed face position are generated and displayed as indexes that represent the sizes of multiple patches corresponding to multiple positions on the setting screen image, that is, the sizes of the learning model application areas, respectively. used when

Next, in S405, the display unit 306 receives an instruction from the area determination unit 303, and superimposes the index representing the assumed face size and assumed face position calculated in S404 on the setting screen. In the present embodiment, an example in which icons are superimposed and displayed as indicators for the assumed face size and the assumed face position will be given.

FIG. 6 is a diagram showing an example of a setting screen on which indexes (icons) corresponding to assumed face sizes are superimposed and displayed.
In FIG. 6, a setting screen 601 shows a screen similar to that in FIG. 5 for accepting input of face size and face position by the user. 604 to 606 show an example of superimposed display. Cursors 602 and 603 are similar to the cursors 502 and 503 described above with reference to FIG. 5, and are cursors for designation when the user designates the face size and face position. Icons 604 to 606 are icons as indices corresponding to the assumed face size and assumed face position. The display unit 306 creates a plurality of icons 604 to 606 based on the information on the assumed face size and assumed face position calculated in S404, and superimposes them on the screen. That is, in FIG. 6, the assumed face size and assumed face position corresponding to a plurality of learning model application areas corresponding to each position on the image are determined by the area determination unit 303, and the assumed face size and assumed face position corresponding to the plurality of assumed face sizes and assumed face positions are determined. This is an example in which a plurality of icons 604 to 606 are displayed.

FIG. 7 is a diagram used to explain display positions when icons are superimposed and displayed.
A frame 701 shown in FIG. 7 represents a patch (learning model application area) for each of a plurality of positions on the image of the setting screen 601 shown in FIG. Cursors 702 and 703 are assumed to be specification cursors for specifying the face size and face position by the user, similar to the cursors 502 and 503 described with reference to FIG. Icons 704 to 706 are created based on the assumed face size and assumed face position calculated for each patch (learning model application area) indicated by the frame 701 . An icon 704 is an icon created based on the assumed face size and assumed face position calculated from the lower patch. Similarly, an icon 705 and an icon 706 are created from the middle patch and the upper patch, respectively, based on the calculated assumed face size and assumed face position. Here, an example is shown in which an icon corresponding to the assumed face size, which is the median value of the size range that the assumed face size can take in the patch at each position, is created.

Also, the positions where the created icons are displayed are the centers of the patches of different sizes that do not overlap the cursors 702 and 703 . That is, the positions at which the created icons are displayed can be set to the assumed face positions closer to the result of the estimation processing by determining the positions based on the patches. At this time, each icon created based on the assumed face size may be arranged by adjusting the position so as to form a vertical straight line in consideration of the user's visibility, or may be arranged at a fixed position on the screen (for example, , right edge of the screen).
In the example of superimposed display shown in FIGS. 6 and 7, an icon is created and displayed from the assumed face size and assumed face position for each patch. good.

FIGS. 8A and 8B are diagrams showing an example of superimposed display of a plurality of icons for each patch. That is, in the case of FIGS. 8A and 8B, the area determination unit 303 determines the size range that the assumed face size can take for the learning model setting area. Then, the display unit 306 displays a plurality of icons as indices representing the possible size range of the assumed face size.

FIG. 8(a) is an example of displaying so that it can be seen that the size values of the assumed face size have a certain range. FIG. 8(a) is an icon image representing an index representing the range of possible size values of the assumed face size, from an index representing the minimum size to an index representing the maximum size within the range of the assumed face size. shows an example of displaying . That is, for example, on the near side of the screen, icon images are displayed that represent from an icon 805 representing the minimum assumed face size that can be taken corresponding to the position to an icon 806 representing the maximum assumed face size. On the other hand, on the far side of the screen, icon images representing an icon 801 representing the minimum assumed face size that can be taken corresponding to the position and an icon 802 representing the maximum assumed face size are displayed. Also, between the front side and the back side of the screen, icon images are displayed that represent from an icon 803 that represents the minimum assumed face size that can be taken corresponding to the position to an icon 804 that represents the maximum assumed face size. be. In this manner, by displaying FIG. 8A in place of the icons 704 to 706 described above, the user can be notified of the possible size range of the assumed face size.

In addition, the icon size is not limited to this display example, and may be displayed by changing the icon size by animation. For example, an animation display that sequentially changes the size and position of the icon display, or an animation display that displays small-sized icons at the back of the screen and displays larger icons at the front of the screen. is mentioned.

Furthermore, FIG. 8(b) is an example of displaying in such a way that people with assumed face sizes are crowded together. FIG. 8(b) includes an index representing the minimum size to the maximum size within the range that the value of the assumed face size can take as an index representing the range that the value of the assumed face size can take. An example of displaying an aggregate image is shown. That is, for example, a group image (crowd image) 810 in which a plurality of persons from the person corresponding to the maximum assumed face size to the person corresponding to the minimum assumed face size gather from the front side to the back side of the screen. is displayed. In this way, instead of the icons 704 to 706 described above, as shown in FIG. A user can be notified of a situation in which people of the same face size are crowded. The collective image 810 may be displayed by displaying an image prepared in advance, or by displaying an image obtained by synthesizing a plurality of person CGs generated by computer graphics (CG). good.

Not limited to the example of FIG. 8, if the user can present information that allows the user to determine the size of the learning model application area in which people are to be estimated when the number of people estimation process is performed, Any method may be used. For example, icons may be displayed on all patches, icons may be translucent, characters may be used instead of icons, or actual images may be used instead of icons.

Returning to the flowchart of FIG. 4 and proceeding to S406, the region determination unit 303 determines whether the setting of the face size and the face position obtained in S402 has been completed. Here, whether or not the setting is completed may be determined by receiving an input from the user. Alternatively, the area determination unit 303 calculates the difference between the face position and face size obtained in S402 and S404 and the assumed face position and assumed face size, and determines whether the difference is equal to or less than a threshold. good too. That is, if the difference between them is equal to or less than the threshold, the value specified by the user (face position/face size) is close to the assumed value (assumed face position/assumed face size) determined by the area determination unit 303. , the setting may be regarded as completed. If it is determined in S406 that the setting has been completed, the process proceeds to S407, and if it is determined that the setting has not been completed, the process returns to S402.

After proceeding to S407, the area determination unit 303 divides the image according to the learning model application area obtained in S403. In this embodiment, these divided images are used as patches.
Next, in S408, the estimating unit 304 receives the result of the learning model application region obtained in S403, and executes the number of people estimation process. The estimating unit 304 resizes the patch image and inputs it to the regressor for estimating the number of people, thereby obtaining the result of estimating the number of people in each region.
Next, in S409, the estimating unit 304 determines whether all divided patch images have been processed, and if the processing has not ended, repeats the processing of S408. On the other hand, if the population estimation process has been executed for all patches, the process proceeds to S410.

After proceeding to S410, the integrating unit 305 receives each result of the population estimation process obtained by the processes of S408 and S409, and integrates those results. In other words, the integration unit 305 calculates the result of estimating the number of people in the entire image by adding up each result. After that, the processing of the flowchart of FIG. 4 ends.

As described above, according to the information processing system of the present embodiment, the face size and face position that can be estimated by the image analysis process are displayed on the setting screen as information necessary for parameter setting for performing the image analysis process. do. This makes it easier for the user to determine whether appropriate analysis settings have been made.

Further, in the case of the information processing system of the present embodiment, execution of image analysis processing is also reduced, so it is possible to reduce the load of image analysis processing. That is, in Patent Document 2, the analysis result is used to present the user with the information necessary to properly perform the setting work and the parameter correction work for obtaining the desired processing result. On the other hand, the information processing apparatus of the present embodiment determines the size of an area that can be estimated by a predetermined estimation process based on the size and position of the area specified by the user, and the determined estimation process The user is presented with an index representing the size of an area that can be estimated. In other words, in the case of Patent Document 2, the user is suggested to correct the parameters using the analysis results. conduct. For this reason, according to the present embodiment, it is possible to provide feedback to the user as to what kind of analysis results are obtained before image analysis processing is executed, and image analysis processing that is less necessary is executed. processing load can be reduced.

The present invention supplies a program that implements one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus reads and executes the program. It can also be realized by processing to It can also be implemented by a circuit (eg, ASIC) that implements one or more functions.
All of the above-described embodiments merely show specific examples for carrying out the present invention, and the technical scope of the present invention should not be construed to be limited by these. That is, the present invention can be embodied in various forms without departing from its technical concept or main features.

302: image acquisition unit, 303: area determination unit, 304: estimation unit, 305: integration unit, 306: display unit

Claims (11)

Determination means for determining the size of an area that can be estimated by a predetermined estimation process based on the size and position of an area specified by a user on a setting screen on which an image acquired by imaging is displayed;
display means for displaying, on the setting screen for the user to specify the size and position of the area, an indicator representing the size of the area that can be estimated by the estimation process determined by the determination means;
An information processing device comprising: The determination means determines a plurality of sizes corresponding to a plurality of positions on the image displayed on the setting screen as sizes of areas that can be estimated by the estimation process,
2. The information processing apparatus according to claim 1, wherein said display means displays a plurality of indicators respectively representing a plurality of sizes corresponding to a plurality of positions on said image determined by said determination means. . The determining means determines a possible size range of an area that can be estimated by the estimation process,
3. The information processing apparatus according to claim 1, wherein said display means displays said index representing said size range. The display means displays, as the index representing the range of sizes, an image representing an index representing a minimum size to an index representing a maximum size within the range of sizes. Item 4. The information processing device according to item 3. The display means displays, as the index representing the range of sizes, a group image including an index representing a minimum size to an index representing a maximum size within the range of sizes. The information processing apparatus according to claim 3. The predetermined estimation process is a process of detecting at least a predetermined target object from the captured image using a learning model,
The determining means determines the size of the area to which the learning model is applied based on the size and position of the area designated by the user on the setting screen,
6. The information processing apparatus according to any one of claims 1 to 5, wherein said display means displays an index representing the size of the area to which said learning model is applied. 7. The information processing apparatus according to claim 6, wherein the predetermined estimation processing includes processing for estimating the number of the detected predetermined objects from the captured image using the learning model. 8. An information processing apparatus according to claim 6, wherein said predetermined object is a person. 9. The information processing apparatus according to claim 8, wherein the size and position of the region designated by the user are the face size and position of the person. An information processing method executed by an information processing device,
a determination step of determining the size of an area that can be estimated by a predetermined estimation process based on the size and position of an area specified by a user on a setting screen displaying an image acquired by imaging;
a display step of displaying, on the setting screen for the user to specify the size and position of the region, an indicator representing the size of the region that can be estimated by the estimation process determined in the determination step;
An information processing method characterized by having A program for causing a computer to function as the information processing apparatus according to any one of claims 1 to 9.

Images