Person tracking assist device

Abstract

Provided is a person tracking assistance device capable of improving the efficiency of a tracking operation when a user tracks a person appearing in an image captured by a camera. The person tracking assistance device recognizes one or more persons appearing in at least one of the plurality of images. The person tracking assistance device displays a person tracking screen on a display device, the person tracking screen including: a first display unit for synchronously playing and displaying a monitoring image selected from a plurality of images during an image time period; and a second display unit that displays at least one appearance section in which one person or a plurality of persons appear in the monitoring image in the image time. When a user performs an operation for specifying one point in time of an appearance interval relating to a target person among one or more persons, the person tracking assistance device displays snapshots of one or more monitoring images at the specified one point in time. People tracking information is displayed on the people tracking screen so that the user recognizes the portion where the target person appears in the snapshot.

Description

Technical Field

[0001] This disclosure relates to techniques for assisting in the tracking of people appearing in images captured by a camera. Background Technology

[0002] Patent Document 1 discloses a crime prevention system for identifying suspicious actions of people captured in camera images. The system monitors the actions of people captured in multiple camera images captured sequentially over time by a surveillance camera. It detects the human regions of each person in the multiple camera images and performs tracking processing to identify the people in the multiple camera images in a time sequence based on the image data of the human regions. Furthermore, if the identification process performed by tracking fails midway, the system determines that a person engaging in suspicious behavior is present in the camera images.

[0003] In addition, as documents representing the technical level in this field, there are the following patent documents 2 and 3.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2023-086471

[0005] Patent Document 2: Japanese Patent Application Publication No. 2023-156963

[0006] Patent Document 3: Japanese Patent Application Publication No. 2024-032346

[0007] Consider a scenario where a user wants to simultaneously review and track a specific person appearing in an image. In this case, person tracking typically spans multiple images captured by multiple cameras. Therefore, the user needs to locate the person to be tracked within each of the multiple images, one by one, over the duration of the image. Historically, such tracking operations have required significant manpower and time resources. Summary of the Invention

[0008] This disclosure was made in view of the aforementioned issues. One object of this disclosure is to provide a technique that can improve the efficiency of tracking operations when a user tracks people appearing in an image.

[0009] One aspect of this disclosure relates to a person tracking assistive device. The person tracking assistive device includes: a storage device for storing data of multiple images captured by multiple cameras; and a processing circuit. The processing circuit identifies one or more people appearing in at least one of the multiple images. Furthermore, the processing circuit displays a person tracking screen on a display device, wherein the person tracking screen includes: a first display unit that displays one or more surveillance images selected from the multiple images in a synchronized playback manner during the image time; and a second display unit that displays the appearance intervals of one or more people appearing in at least one of the surveillance images during the image time. Moreover, when a user specifies a time point in the appearance interval related to an object among the one or more people on the person tracking screen, the processing circuit displays a snapshot of one or more surveillance images at that time point on the first display unit, and displays person tracking information on the person tracking screen that allows the user to identify the portion of the object appearing in the snapshot.

[0010] According to this disclosure, a person tracking screen is displayed, comprising: a first display unit that displays one or more selected surveillance images in a synchronized playback manner during the image time; and a second display unit that displays the appearance interval of at least one of the surveillance images during the image time for each identified person or person. Thus, the user can easily determine which interval each person appears in the image. Furthermore, if the user specifies a time point within the appearance interval related to the target person, a snapshot of one or more surveillance images at that time point and person tracking information are displayed. Thus, the user can easily identify the portion of the target person appearing at each time point between appearance intervals. Therefore, according to this disclosure, the efficiency of user-performed tracking operations can be improved. Attached Figure Description

[0011] Figure 1 This is a conceptual diagram used to illustrate the outline of the human tracking assistance device involved in the implementation.

[0012] Figure 2 This is a block diagram illustrating the basic structure of the human tracking assistance device involved in the implementation method.

[0013] Figure 3 This is a block diagram illustrating the functional structure associated with the processing performed by the person tracking assistance device involved in the implementation.

[0014] Figure 4 It means by Figure 3The diagram shows an example of human identification information generated by the human identification unit.

[0015] Figure 5 This is an example diagram illustrating the structure of a screen where a person is tracking the image.

[0016] Figure 6 This diagram illustrates the user's actions when tracking an object or person in a person tracking screen.

[0017] Figure 7 This is a diagram representing the processing sequence when a user tracks an object or person in a person tracking screen.

[0018] Figure 8 It is a concept diagram used to illustrate the display of images in a hypothetical space.

[0019] Explanation of reference numerals in the attached figures:

[0020] 2…User; 20…Camera; 30…Image; 40…Imaginary space image; 100…Person tracking assist device; 110…Processing circuit; 120…Storage device; 121…Computer program; 200…User interface; 210…Display device; 220…Input device; 300…Person tracking screen; 310…First display unit; 320…Second display unit; 322…Appearance range; 323…Person tracking information; BB…Bounding box; CF…Camera information; D10…Image data. Detailed Implementation

[0021] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. Furthermore, for structures that are the same or equivalent in the various figures, the same reference numerals will be used, and their descriptions will be simplified or omitted.

[0022] 1-person tracking assist device

[0023] Figure 1 This is a conceptual diagram illustrating the outline of a person tracking assist device 100. The person tracking assist device 100 acquires multiple images 30 captured by multiple cameras 20. Each camera 20 is configured to monitor a person 10 moving within a defined area. Examples of defined areas include streets, buildings, etc. The person 10 can also be referred to as a "pedestrian 10". The multiple cameras 20 may also include two or more cameras capturing images of the same location from different viewpoints.

[0024] The person tracking assist device 100 uses data from multiple acquired images 30 to display a "person tracking screen 300" on the display device 210. This "person tracking screen 300" assists the user 2 in tracking the person 10 appearing in the images 30. The user 2 tracks the person 10 appearing in the images 30 on the person tracking screen 300. The person tracking screen 300 will then be described in detail. Tracking the person 10 appearing in the images 30 is useful in fields such as traffic management and action analysis.

[0025] Here, we examine the scenario where User 2 wants to track a specific person 10-T (hereinafter referred to as "object person 10-T"). In this case, User 2 needs to locate the object person 10-T one by one during the time period of multiple images 30. Figure 1 In the example shown, frames at time points t1, t2, and t3 are displayed sequentially according to the three images 30 (30-A, 30-B, and 30-C). Figure 1 In the example shown, user 2 confirms that person 10-T is reflected in image 30-A between time point t1 and time point t2, and then tracks person 10-T. Similarly, user 2 confirms that person 10-T is reflected in image 30-B between time point t2 and time point t3, and confirms that person 10-T is reflected in image 30-C at time point t3, thus tracking person 10-T. Therefore, in the tracking operation of person 10-T appearing in image 30, user 2 needs to confirm in which time interval and in which camera 20's image 30 person 10-T is reflected. In particular, when tracking person 10-T across multiple images 30, it is necessary to identify the person identical to person 10-T from the people 10 appearing in different images 30. Previously, such tracking operations performed by user 2 required significant manpower or time costs. This embodiment proposes a human tracking assistance device 100 that displays a human tracking screen 300, enabling user 2 to perform tracking operations efficiently.

[0026] Figure 2This is a block diagram illustrating the basic structure of the human tracking assist device according to this embodiment. The human tracking assist device 100 is a computer equipped with a processing circuit 110 and one or more storage devices 120 (hereinafter referred to as "storage devices 120"). The human tracking assist device 100 is connected to multiple cameras 20 via a communication network 3. The communication network 3 is, for example, a wired communication system, a wireless communication system, or the Internet. Furthermore, the human tracking assist device 100 is connected to a user interface 200, which includes a display device 210 and an input device 220. The display device 210 displays various information based on the output from the human tracking assist device 100. In particular, the display device 210 displays a human tracking screen 300. The display device 210 is, for example, a monitor, a projector, or a head-mounted display (HMD). The input device 220 receives various inputs from the user 2 to the human tracking assist device 100. The input device 220 is, for example, a keyboard, a pointing device, a touchpad, a switch, or a microphone. The person tracking assist device 100 can also be a server accessible via the communication network 3. In this case, the user interface 200 can also be a user terminal (e.g., a personal computer, smartphone, tablet, etc.) connected to the person tracking assist device 100 via the communication network 3.

[0027] The processing circuit 110 performs various processes. The processing circuit 110 may be composed of, for example, a general-purpose processor, a purpose-specific processor, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), a FPGA (Field-Programmable Gate Array), an integrated circuit, existing types of circuits, and one or more combinations thereof. A processor including transistors and other circuits is one example of the processing circuit 110. The processing circuit 110 may also be referred to as a circuitry or a processing circuitry. A circuitry is hardware programmed to implement or perform the functions described in this disclosure.

[0028] Storage device 120 stores various information required for the processing of the execution processing circuit 110. Storage device 120 may be composed of recording media such as RAM (Random Access Memory), ROM (Read Only Memory), SSD (Solid State Drive), or HDD (Hard Disk Drive). Storage device 120 stores a computer program 121 executable by the processing circuit 110. Computer program 121 consists of multiple instructions describing the processing performed by the processing circuit 110. Computer program 121 may be recorded on a computer-readable recording medium. The function of human tracking assistance device 100 is realized through the cooperation of processing circuit 110 executing computer program 121 and storage device 120.

[0029] The storage device 120 also stores image data D10. Image data D10 is data used to manage multiple images 30 captured by multiple cameras 20. The human tracking assist device 100 acquires the images 30 captured by each camera 20 and records the acquired images 30 in image data D10 for management. The human tracking assist device 100 can also acquire images 30 from each camera 20 sequentially online. In addition, the human tracking assist device 100 can also acquire images 30 from the user 2 via input device 220. Each image 30 managed by image data D10 includes information about the shooting time and information about the camera 20 that captured the image 30.

[0030] Figure 3 This is a block diagram illustrating the functional structure associated with the processing performed by the human tracking assist device 100. As functional modules, the human tracking assist device 100 includes a human recognition unit F10 and a human tracking screen display unit F20. These functional modules are implemented through the cooperation of the processing circuit 110, which executes the computer program 121, and the storage device 120.

[0031] The human recognition unit F10 performs recognition processing to identify one or more persons 10 appearing in at least one of multiple images 30 managed by image data D10, and generates human recognition information IDF. By performing the recognition processing, each identified person 10 is assigned unique identification information (ID). Furthermore, by performing the recognition processing, the portion of each identified person 10 appearing in each image 30 is determined. The portion of a person 10 appearing in an image 30 is represented, for example, by a bounding box surrounding the person 10. That is, in this case, the position of the bounding box is determined by performing the recognition processing. The portion of a person 10 appearing in an image 30 can also be represented by other methods such as a polygon surrounding the person 10, the skeletal position of the person 10, etc. In the following description, the case where the position of the bounding box is determined by performing the recognition processing is considered.

[0032] Typically, the recognition process consists of tracking processing, which detects and tracks the detected person 10 within the image 30, and human re-identification processing, which identifies the same person 10 across different images 30. Tracking processing and human re-identification processing are well-known techniques, and their methods are not particularly limited. For example, the recognition process can be implemented through an AI pipeline that includes a learned machine learning model configured to perform tracking processing and human re-identification processing on multiple images 30 as input.

[0033] The human identification information (IDF) is information about each of the multiple images 30, which establishes a correlation between the ID of the person 10 projected in each frame of the image 30 and the portion of the image 30 in which the person 10 with that ID appears. Figure 4 This is a diagram illustrating an example of human identification information (IDF) associated with one of multiple images 30. Figure 4 In the example shown, for each frame from frame 0 to frame 3, the ID of person 10 projected in image 30 is displayed in association with the position of the bounding box surrounding that ID in image 30. The position of the bounding box is defined by the positions of its left and top edges, and its width and height values. Figure 4 As shown in the examples, person 10 with the ID "8aff82a2" is projected in image 30 within the frame interval from frame 0 to frame 3. Similarly, person 10 with the ID "50ca6660" is projected in image 30 within the frame interval from frame 1 to frame 3. In particular, the positions of the bounding boxes surrounding these persons 10 in each frame can be determined.

[0034] By referring to the human identification information IDF generated by the human identification unit F10, for each of the multiple images 30, it is possible to obtain the range in which the identified person 10 is reflected in the image 30 and the portion of the image 30 in which the person appears. The human identification unit F10 may also store the generated human identification information IDF in the storage device 120.

[0035] Refer again Figure 3 The human tracking screen display unit F20 generates a human tracking screen 300 based on multiple images 30 managed by image data D10 and human identification information IDF obtained from the human recognition unit F10, and displays it on the display device 210. Furthermore, the human tracking screen display unit F20 receives operation input from the user 2 on the human tracking screen 300 displayed on the display device 210 via the input device 220. The human tracking screen display unit F20 changes the display of the human tracking screen 300 according to the operation input from the user 2.

[0036] The following is a detailed description of the human tracking screen 300 displayed on the display device 210 by the human tracking screen display unit F20.

[0037] Two people tracking footage

[0038] 2.1 Screen Structure

[0039] Figure 5 This is a concept diagram used to explain the screen structure of the human tracking screen 300. The human tracking screen 300 includes a first display unit 310 and a second display unit 320.

[0040] The first display unit 310 displays one or more images 30 simultaneously during the image duration. The one or more images 30 displayed on the first display unit 310 are selected by the user 2 from multiple images 30 managed by image data D10. The human tracking screen display unit F20 accepts the selection input of the image 30 from the user 2. Selection input can also be made by selecting a camera 20. For example, the user 2 selects the camera 20 from a list of multiple cameras 20 from which they want to view the image 30. In this case, the image 30 captured by the selected camera 20 is displayed on the first display unit 310. The human tracking screen display unit F20 can also further accept the input of specifying the image duration from the user 2. For example, the user 2 specifies a period based on the shooting date and time. Hereinafter, the one or more images 30 displayed on the first display unit 310 will also be referred to as "monitoring images 30". Figure 5 In the example shown, as monitoring image 30, two images 30-X and 30-Y are shown.

[0041] The first display unit 310 displays a button 311 for starting playback of the monitoring video 30 and a progress bar 312 indicating the playback position of the monitoring video 30 within the video duration. If the user 2 operates the button 311, "synchronized playback" of the monitoring video 30 begins. That is, each monitoring video 30 is played on the first display unit while synchronizing the shooting times. Alternatively, playback of the monitoring video 30 can be temporarily paused when the user 2 operates the button 311 during playback. The progress bar 312 moves according to the playback position of the monitoring video 30. By checking the position of the progress bar 312, the user 2 can determine the playback position of the monitoring video 30 within the video duration.

[0042] In each surveillance image 30, the bounding box BB surrounding the identified person 10 is displayed in an overlapping manner. Figure 5 In each of images 30-X and 30-Y, except for one, the reference numerals for person 10 and bounding box BB are omitted. The person tracking display unit F20 displays the bounding box BB in each frame of each monitoring image 30 by referring to the person identification information IDF. As a variation of this embodiment, the bounding box BB can also be replaced by the person identification information IDF, and the polygon surrounding person 10 and the skeletal position of person 10 can be displayed in each monitoring image 30 in an overlapping manner.

[0043] In the first display unit 310, camera information CF of the camera 20 that captures the surveillance image 30 is displayed in association with the surveillance image 30. The camera information CF is information that identifies the camera 20. Figure 5 In the example shown, the camera information CF is an identification number unique to each camera 20. The human tracking display unit F20 displays the camera information CF by referring to the camera information CF contained in each surveillance image 30. The user 2 can confirm the configuration and specification information of the camera 20 that captured each surveillance image 30 by referring to the management data of the camera 20 based on the camera information CF. Furthermore, the camera information CF is information unique to each surveillance image 30. Therefore, each surveillance image 30 can be identified and distinguished based on the camera information CF. Figure 5 In the example shown, the camera information CF-X of the camera 20 capturing image 30-X and the camera information CF-Y of the camera 20 capturing image 30-Y are displayed in association with image 30-X and image 30-Y, respectively. As a variation of this embodiment, other information that can identify and distinguish each monitored image 30 may be displayed instead of the camera information CF. For example, numbers indicating the display order of the monitored images 30 and identification numbers inherent to each monitored image 30 may also be displayed.

[0044] The second display unit 320 displays the interval 322 (hereinafter referred to as "appearance interval 322") in which the identified one or more persons 10 are reflected in at least one of one or more surveillance images 30 during the image time. The person tracking screen display unit F20 obtains the appearance interval 322 of each person 10 identified relative to the surveillance image 30 by referring to the person identification information IDF.

[0045] exist Figure 5 In the example shown, the IDs 321 of each identified person 10 are displayed in a list on the second display unit 320. Furthermore, the second display unit 320 displays the occurrence intervals 322 of each person 10 in association with their IDs 321. Additionally, in... Figure 5 Except for a few parts, the reference numerals for ID321 and appearance interval 322 are omitted. By confirming the appearance interval 322 associated with ID321, user 2 can easily determine the time point in the monitoring image 30 when person 10 with ID321 appears in the image time.

[0046] 2.2 Tracking of people in the scene

[0047] In the person tracking screen 300 of the above-described screen structure, user 2 can track the object person 10-T as follows. Figure 6 This is a concept diagram used to illustrate the tracking of the object 10-T in the human tracking screen 300.

[0048] First, user 2 performs an operation that specifies a point in time within the occurrence interval 322 related to the object 10-T. For example, user 2 operates the pointing device, causing the mouse pointer 4 to hover over a point in time within the occurrence interval 322. Figure 6 In the example shown, person 10 with ID 321, “8aff82a2”, is the case of person 10-T.

[0049] If user 2 specifies a time point within the occurrence interval 322, a snapshot of each monitoring image 30 at the specified time point is displayed on the first display unit. That is, each monitoring image 30 stops at the playback position corresponding to the specified time point. The progress bar 312 can also be moved to the playback position corresponding to the specified time point. A snapshot can also be referred to as a "still image."

[0050] If user 2 further specifies a time point within the occurrence interval 322, "human tracking information 323" is displayed on the human tracking screen 300. Human tracking information 323 is information that allows user 2 to identify the portion of the subject person 10-T appearing in the snapshot. In particular, in this embodiment, human tracking information 323 is information that establishes a correlation between the image IM of the region surrounding the bounding box BB of the subject person 10-T cropped from the snapshot and the camera information CF of the camera 20 that captured the surveillance image 30 of the subject person 10-T appearing.

[0051] User 2 can identify the portion of the target person 10-T appearing in a snapshot of the surveillance image 30 based on the person tracking information 323 involved in this embodiment, as follows. User 2 can determine the surveillance image 30 (target surveillance image) where the target person 10-T appears by comparing the camera information CF of the person tracking information 323 with the camera information CF associated with each surveillance image 30. In this sense, the camera information CF of the person tracking information 323 can be considered as information for determining the target surveillance image. Then, User 2 can identify the portion of the target person 10-T appearing by comparing the image IM of the person tracking information 323 with the image on the target surveillance image surrounded by each bounding box BB. Thus, User 2 can easily identify the portion of the target person 10-T appearing in a snapshot of the surveillance image 30 based on the person tracking information 323. To make it easier to identify the part of the object 10-T that appears, the person tracking display unit F20 can also make the color of the bounding box BB surrounding each person 10 consistent with the color of the appearance range 322 related to each person 10.

[0052] By appropriately changing the specified time points within the appearance interval 322 related to the object 10-T, user 2 can easily identify the portions of the object 10-T that appear at each time point within the appearance interval 322. Thus, according to this embodiment, user 2 can efficiently track the object 10-T in the person tracking screen 300.

[0053] Figure 7 This is a diagram showing the processing sequence when user 2 is tracking object 10-T in person tracking screen 300.

[0054] First, in step S10, user 2 operates input device 220 to specify a time point in the occurrence interval 322 related to the object person 10-T. The operation input by user 2 is sent to person tracking assist device 100.

[0055] If a time point in interval 322 has been specified, then in step S20, the human tracking display unit F20 displays a snapshot of the monitoring image 30 at the specified time point in the first display unit 310.

[0056] Next, in step S30, the person tracking display unit F20 generates person tracking information 323 for the target person 10-T based on the person identification information IDF. More specifically, the person tracking display unit F20 cuts out the region of the bounding box BB surrounding the target person 10-T from the snapshot of the monitoring image 30. Then, the person tracking display unit F20 establishes a correlation between the image IM of the cut-out region surrounding the target person 10-T and the camera information CF of the monitoring image of the target person 10-T.

[0057] Next, in step S40, the person tracking screen display unit F20 displays the generated person tracking information 323 on the person tracking screen 300.

[0058] 2.3 Display of Imaginary Space Images

[0059] To make the tracking of the object 10-T performed by user 2 more efficient, the person tracking screen display unit F20 can further display "imaginary space image" on the person tracking screen 300. The imaginary space image is an image of the movement of person 10 appearing in the monitoring image 30 during the image time period, represented by a target in imaginary space.

[0060] Figure 8 This is a conceptual diagram used to explain the display of the virtual space image 40. The human tracking display unit F20 represents the movement of the person 10 appearing in the monitoring image 30 through a target OB on the virtual space VS. This can be achieved, for example, by applying known image analysis techniques to the monitoring image 30 to infer the 3D pose of the person 10, and by reflecting the inferred 3D pose onto the target OB. The virtual space VS can be pre-generated. The target OB can also be a virtual avatar from the internet. The human tracking display unit F20 generates the virtual space image 40 that projects the virtual space from a predetermined viewpoint. Then, the human tracking display unit F20 displays the generated virtual space image 40 on the human tracking screen 300 in a manner synchronized with the monitoring image 30. Figure 8 In the example shown, the virtual space image 40 is displayed on the first display unit 310. If the user 2 operates the button 311, the virtual space image 40 will begin to play synchronously with the monitoring image 30. The viewpoint of the virtual space image 40 can also be changed arbitrarily by the user 2.

[0061] By displaying the virtual space image 40 in this way, user 2 can confirm the movements of the object 10-T from various viewpoints through the virtual space image 40. As a result, user 2 can track the object 10-T more efficiently.

[0062] 3 Effects

[0063] As explained above, according to this embodiment, a person tracking screen can be displayed, which includes: a first display unit 310 that displays one or more surveillance images 30 in a synchronized manner during the image time; and a second display unit 320 that displays the appearance interval 322 of at least one surveillance image of one or more surveillance images 30 during the image time for each identified person or person 10. Thus, the user can easily determine in which interval each person 10 appearing in the image 30 is reflected. Furthermore, if the user 2 performs an operation to specify a time point of the appearance interval 322 related to the target person 10-T, a snapshot of one or more surveillance images 30 at that time point and person tracking information 323 are displayed. Thus, the user can easily identify the portion of the target person 10-T appearing at each time point between the appearance intervals 322. Therefore, according to this embodiment, the efficiency of the tracking operation performed by the user 2 can be improved.

Claims

1. A human tracking auxiliary device, characterized in that, have: A storage device that stores data of multiple images captured by multiple cameras; and Processing circuit, The processing circuit is configured as follows: Identify one or more persons appearing in at least one of the plurality of images, and The person tracking footage is displayed on the display device, where... The person tracking footage includes: A first display unit, which displays one or more selected surveillance images from the plurality of images in a synchronized playback manner during the image duration; and The second display unit displays the appearance intervals of at least one of the one or more surveillance images during the image time, where the one or more persons appear in the one or more surveillance images. When a user performs an operation on the person tracking screen to specify a point in time within the occurrence interval related to the object among the one or more people, The processing circuit displays snapshots of one or more surveillance images at one time point in the first display unit, and The person tracking information that allows the user to identify portions of the person appearing in the snapshot will be displayed on the person tracking screen.

2. The human tracking auxiliary device according to claim 1, characterized in that, The processing circuit is further configured as follows: In the one or more surveillance images displayed on the first display unit, the bounding boxes surrounding the one or more people are displayed in an overlapping manner. The person tracking information is information that is associated with an image of the region surrounding the bounding box cut out from the snapshot and information that identifies the appearance of the person in one or more surveillance images.

3. The human tracking auxiliary device according to claim 2, characterized in that, The information used to determine the object surveillance image is the information of the camera among the plurality of cameras that captured the object surveillance image.

4. The human tracking auxiliary device according to claim 1, characterized in that, The processing circuit is further configured as follows: Generate an imaginary spatial image representing the actions of one or more people during the time period of the image through targets in the imaginary space, and The hypothetical space image is displayed on the person tracking screen in a manner that is synchronized with one or more surveillance images.

5. A human tracking auxiliary device, characterized in that, have: A storage device that stores data of multiple images captured by multiple cameras; and Processing circuit, The processing circuit is configured as follows: For each of the multiple images, the people within the images are identified. The region surrounding the person is cropped from the snapshot of the image where the person appears, and Generate an image of the region surrounding the person cut out from the snapshot, and establish person tracking information associated with information from the cameras among the plurality of cameras that captured images of the person.

Images