Anthropometric data acquisition method, processing method and device

Abstract

This application provides a method, processing method, and device for acquiring anthropometry data, relating to the field of computer vision. The acquisition method includes: a terminal device acquiring key point data of a target person in a real-time image of a shooting interface; determining whether to push a prompt message based on the key point data; the prompt message prompting the target person to adjust their standing position and / or posture. After the target person adjusts their position and posture according to the prompt message, the terminal device acquires a full-body image of the target person whose standing position and posture both meet preset requirements, and determines the target person's anthropometry data based on the key point data in the full-body image. The above process involves posture recognition of the person's key point data. Posture recognition determines whether a shooting prompt is needed during the shooting process, thereby guiding the person to acquire full-body front and side images, achieving rapid acquisition of images that meet the standing posture requirements, and improving the accuracy of anthropometry data.

Description

Technical Field

[0001] This application relates to the field of computer vision technology, and in particular to a method, processing method and device for acquiring human body measurement data. Background Technology

[0002] With the rapid development of information technology, the apparel industry is also moving towards digitalization and intelligentization. To meet users' needs for online clothing purchases, e-commerce platforms have launched intelligent body measurement mini-programs, making it convenient for users to take measurements and select clothing online.

[0003] To improve the accuracy of body measurements, the full-body photos taken by users need to meet preset standing posture requirements. Existing smart body measurement mini-programs cannot judge or guide the user's standing posture during the photo shoot; users must manually confirm the correctness of their posture in the photo after taking it. The accuracy of calculating anthropometric data based on user-confirmed photos is not high. Summary of the Invention

[0004] This application provides a method, processing method, and device for acquiring and processing anthropometric data. By judging and guiding the posture of people in images, high-quality images of people's postures are acquired, thereby improving the accuracy of anthropometric data.

[0005] A first aspect of this application provides a method for acquiring human body measurement data, applied to a terminal device, wherein the terminal device displays a real-time image on its camera interface; the method includes:

[0006] Acquire key point data of the target person in the real-time image;

[0007] By analyzing the key point data of the target person, it is determined whether to push a prompt message. The prompt message is used to prompt the target person to adjust their standing position and / or posture.

[0008] A full-body image of the target person is acquired, provided that the standing position and posture meet the preset requirements. Based on the key point data of the target person in the full-body image, the anthropometric data of the target person is determined. The anthropometric data is used to indicate the measurement data of each part of the target person's body.

[0009] In an optional embodiment of the first aspect of this application, obtaining key point data of the target person in the real-time image includes:

[0010] Based on a lightweight human keypoint detection model, keypoint data of the target person is obtained from the real-time image. The keypoint data includes the position data of all keypoints of the target person.

[0011] In an optional embodiment of the first aspect of this application, determining whether to push a notification message by analyzing the key point data of the target person includes:

[0012] The number of key points is obtained based on the key point data of the target person;

[0013] If the number of key points is less than a preset key point threshold, the first instruction information is pushed. The first instruction information is used to instruct the target person to stand in a designated area of ​​the shooting interface.

[0014] In an optional embodiment of the first aspect of this application, determining whether to push a notification message by analyzing the key point data of the target person includes:

[0015] Determine the detection box corresponding to the target person based on the key point data of the target person;

[0016] Obtain the area ratio of the detection box to the real-time image;

[0017] If the area ratio is less than a preset area ratio threshold, a second prompt message is pushed. The second prompt message is used to instruct the target person to move forward to a designated area of ​​the shooting interface.

[0018] In an optional embodiment of the first aspect of this application, determining whether to push a notification message by analyzing the key point data of the target person includes:

[0019] The posture data of the target person is determined based on the key point data of the target person;

[0020] If the posture data is outside the preset value range corresponding to the posture data, a third prompt message is pushed. The third prompt message is used to instruct the target person to adjust their standing posture.

[0021] In an optional embodiment of the first aspect of this application, the attitude data includes at least one of the following:

[0022] Upper limb opening and closing degree, which is determined based on the position data of key points of the upper limbs of the target person;

[0023] Shoulder offset, which is determined based on the position data of the key points of the left and right shoulders of the target person;

[0024] The torso offset is determined based on the position data of the key points of the target person's limbs and height.

[0025] Lower limb curvature, which is determined based on the location data of key points of the target person's lower limbs and height;

[0026] Upper limb verticality, which is determined based on the positional data of the key points of the upper limbs and the key points of the buttocks of the target person.

[0027] In an optional embodiment of the first aspect of this application, the method further includes:

[0028] If the notification message is to be pushed, the notification message will be displayed on the camera interface of the terminal device, or the notification message will be broadcast through the speaker of the terminal device.

[0029] In an optional embodiment of the first aspect of this application, the full-body image includes a frontal full-body image and a side full-body image of the target person; the step of determining the anthropometric data of the target person based on the key point data of the target person in the full-body image includes:

[0030] Acquire the first key point data of the target person in the frontal full-body image and the second key point data of the target person in the side full-body image;

[0031] Based on the first key point data and the second key point data, the anthropometric data of the target person are determined.

[0032] In an optional embodiment of the first aspect of this application, the method further includes:

[0033] The anthropometric data of the target person is stored in the terminal device.

[0034] In an optional embodiment of the first aspect of this application, the method further includes:

[0035] In response to a query operation performed on the graphical user interface of the terminal device, the standard size data of the target item is obtained from the server;

[0036] The graphical user interface includes the target item viewed by the target person, and the query operation is used to request the recommended size of the target item that matches the target person;

[0037] Based on the anthropometric data of the target person and the standard size data, the recommended size is determined and displayed on the graphical user interface.

[0038] A second aspect of this application provides a method for processing anthropometric data, applied to a terminal device, wherein the terminal device displays a target item currently being viewed by a target person on an item interface, and the method includes:

[0039] When a trigger operation on the processing control of the item interface is detected, the standard size data of the target item is obtained from the server;

[0040] If the terminal device does not include the anthropometric data of the target person, the system will enter the shooting interface and acquire a full-body image of the target person whose standing position and posture meet the preset requirements. The shooting interface includes prompt information, which is used to prompt the target person to adjust their standing position and / or posture.

[0041] By acquiring key point data of the target person in the full-body image, the anthropometric data of the target person is determined, and the recommended size is determined based on the anthropometric data of the target person and the standard size data.

[0042] In an optional embodiment of the second aspect of this application, if the processing control is a control for adding items to a shopping cart, the method further includes:

[0043] After determining the recommended size, add-on item information for the target person is generated based on the recommended size and the identifier of the target item.

[0044] In an optional embodiment of the second aspect of this application, if the processing control is a size query control, the method further includes:

[0045] After determining the recommended size, the system enters the body measurement display interface, which displays the anthropometric data of the target person and the recommended size of the target item.

[0046] A third aspect of this application provides an apparatus for acquiring human body measurement data, comprising:

[0047] The acquisition module is used to acquire key point data of the target person in the real-time image of the terminal device's shooting interface;

[0048] The processing module is used to determine whether to push a prompt message by analyzing the key point data of the target person. The prompt message is used to prompt the target person to adjust their standing position and / or posture.

[0049] The acquisition module is also used to acquire a full-body image of the target person whose standing position and posture both meet the preset requirements;

[0050] The processing module is further configured to determine the anthropometric data of the target person based on the key point data of the target person in the full-body image, wherein the anthropometric data is used to indicate the measurement data of each part of the target person's body.

[0051] A fourth aspect of this application provides a processing apparatus based on human body measurement data, comprising:

[0052] The acquisition module is used to acquire the standard size data of the target item from the server when it detects a trigger operation on the processing control of the item interface of the terminal device.

[0053] The processing module is used to control the entry into the shooting interface if the terminal device does not include the anthropometric data of the target person. The acquisition module is also used to acquire a full-body image of the target person whose standing position and posture meet the preset requirements in the shooting interface. The shooting interface includes prompt information, which is used to prompt the target person to adjust their standing position and / or posture.

[0054] The processing module is further configured to determine the anthropometric data of the target person by acquiring key point data of the target person in the full-body image, and determine the recommended size based on the anthropometric data of the target person and the standard size data.

[0055] A fifth aspect of this application provides an electronic device, including: a memory, a processor, and a computer program; the computer program is stored in the memory and configured to be executed by the processor to implement the method as described in any of the first aspects of this application, or the method as described in any of the second aspects of this application.

[0056] A sixth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method as described in any of the first aspects of this application, or the method as described in any of the second aspects of this application.

[0057] A seventh aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the first aspects of this application, or the method described in any of the second aspects of this application.

[0058] This application provides a method for acquiring anthropometry data. The method includes: a terminal device acquiring key point data of a target person in a real-time image from a shooting interface; determining whether to push a prompt message based on the key point data; the prompt message prompting the target person to adjust their standing position and / or posture. After the target person adjusts their standing position and posture according to the prompt message, the terminal device acquires a full-body image of the target person whose standing position and posture both meet preset requirements; and determines the target person's anthropometry data based on the key point data in the full-body image. The above process involves posture recognition of the person's key point data. By recognizing the posture, it determines whether a shooting prompt is needed during the shooting process, thereby guiding the target person to acquire full-body front and side images, achieving rapid acquisition of images that meet the standing posture requirements, and improving the accuracy of the anthropometry data.

[0059] This application provides a processing method based on human body measurement data. The method obtains the standard size data of the target item based on the target person's operation on the item interface. If the terminal device does not have the target person's human body measurement data pre-stored, the terminal device can be controlled to turn on the camera and obtain the target person's human body measurement data based on the above-mentioned method for obtaining human body measurement data. Then, based on the target person's human body measurement data and the standard size data of the target item, the method can display recommended sizes to the target person or directly add items to their shopping cart, thereby improving the user experience. Attached Figure Description

[0060] Figure 1 Shooting scenario for the anthropometric data acquisition method provided in the embodiments of this application. Figure 1 ;

[0061] Figure 2 Shooting scenario for the anthropometric data acquisition method provided in the embodiments of this application. Figure 2 ;

[0062] Figure 3 A flowchart of a method for acquiring human body measurement data provided in an embodiment of this application;

[0063] Figure 4 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 1 ;

[0064] Figure 5 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 2 ;

[0065] Figure 6 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 3 ;

[0066] Figure 7 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 4 ;

[0067] Figure 8 Schematic diagram of key points in the full-body frontal image of a person provided in the embodiments of this application. Figure 1 ;

[0068] Figure 9 Schematic diagram of key points in the full-body frontal image of a person provided in the embodiments of this application. Figure 2 ;

[0069] Figure 10 A schematic diagram of the pose data of a full-body frontal image of a person provided in an embodiment of this application;

[0070] Figure 11A schematic diagram of pose data for a full-body side view image of a person provided in an embodiment of this application;

[0071] Figure 12 A flowchart illustrating the processing method based on anthropometric data provided in this application embodiment;

[0072] Figure 13 A schematic diagram illustrating the changes in the graphical user interface of the terminal device provided in the embodiments of this application;

[0073] Figure 14 Schematic diagram of the structure of the human body measurement data acquisition device provided in the embodiments of this application Figure 1 ;

[0074] Figure 15 Schematic diagram of the structure of the human body measurement data acquisition device provided in the embodiments of this application Figure 2 ;

[0075] Figure 16 A schematic diagram of the structure of the human body measurement data processing device provided in the embodiments of this application;

[0076] Figure 17 This is a hardware structure diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0078] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in a sequence other than that illustrated or described herein.

[0079] It should be understood that the terms “comprising” and “having” as used herein, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units that are expressly listed, but may include other steps or units that are not expressly listed or that are inherent to such process, method, product, or apparatus.

[0080] In the description of the embodiments of this application, the term "correspondence" may indicate that there is a direct or indirect correspondence between two things, or that there is an association between two things, or that there is a relationship of instruction and being instructed, configuration and being configured, etc.

[0081] Before introducing the method for acquiring human body measurement data provided in this application, we will first briefly introduce the shooting scenarios for this method.

[0082] Figure 1 Shooting scenario for the anthropometric data acquisition method provided in the embodiments of this application. Figure 1 . Figure 1 In the shooting scenario shown, user A is the photographer, and user B is the subject. User A and user B maintain a preset distance, such as 2 meters. User A holds terminal device 11 and uses the external camera of terminal device 11 to capture full-body frontal and full-body side images of user B.

[0083] Figure 2 Shooting scenario for the anthropometric data acquisition method provided in the embodiments of this application. Figure 2 . Figure 2 In the shooting scenario shown, user B is both the photographer and the subject. User B can use an adjustable stand to fix the terminal device 11 at a suitable height, or fix the terminal device 11 on a surface such as a table. The front-facing camera of the terminal device 11 captures a full-body frontal image and a full-body side image of user B.

[0084] Based on any of the above shooting scenarios, terminal device 11 obtains user B's anthropometric data by analyzing user B's full-body frontal and full-body side images. The anthropometric data includes, but is not limited to, the three measurements (a collective term for chest circumference, waist circumference, and hip circumference), shoulder width, neck circumference, arm length, and leg length.

[0085] Currently, in some intelligent body measurement mini-programs, after users take a full-body frontal or full-body side view image, they need to confirm their position and posture according to the prompts to ensure that the captured image can be used for anthropometric data calculation. For example, users need to confirm whether the person is complete in the full-body frontal image, whether the neck is clearly visible, and whether the legs are separated. Similarly, users need to confirm whether the person is complete in the full-body side view image, whether the neck is clearly visible, and whether the front and back contours of the body are clearly visible.

[0086] In summary, existing smart body measurement mini-programs cannot provide real-time judgment and guidance on the user's standing posture during photo shooting. Users need to manually confirm various details after taking the photo. For incorrect shots, users need to take multiple shots, resulting in a poor user experience. Currently, the accuracy of calculating anthropometric data based on user-confirmed photos is not high.

[0087] To address the aforementioned problems, this application provides a method for acquiring anthropometric data. After acquiring real-time images, a lightweight human keypoint detection model pre-installed on the terminal device is first used to detect whether the image contains a person, the number of skeletal keypoints (hereinafter referred to as keypoints), and their locations. Based on the detected keypoint data, the pre-set algorithm in this application is used to identify whether the person's posture meets the requirements for standing. Based on the identification results, it is determined whether to push voice or text prompts, thereby guiding the user to acquire full-body front and side images, achieving rapid acquisition of images that meet standing posture requirements and improving the accuracy of anthropometric data.

[0088] It should be noted that the technical solutions provided in the embodiments of this application are mainly based on Figure 2 The shooting scene shown is the user-controlled shooting scene. Users can adjust their position and posture in real time by referring to the text prompts on the terminal device's shooting interface or the voice prompts from the terminal device's speaker. The shooting process can be completed efficiently and quickly without clicking the screen.

[0089] The technical solutions provided by the embodiments of this application will be described in detail below through specific examples. It should be noted that the technical solutions provided by the embodiments of this application may include some or all of the following contents. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0090] Figure 3 This is a flowchart illustrating a method for acquiring human body measurement data provided in an embodiment of this application. The method for acquiring human body measurement data provided in this embodiment can be applied to terminal devices, such as... Figure 3 As shown, the method includes the following steps:

[0091] Step 101: Obtain key point data of the target person in the real-time image.

[0092] In this embodiment, the terminal device has a pre-installed lightweight human keypoint detection model, which can be used to detect the location of keypoints of a person in an image. Due to the limited memory and computing power of the terminal device, a lightweight model structure should be selected. After model pruning and quantization compression, the model is pre-installed on the terminal device while ensuring detection accuracy. Optional lightweight model structures include, for example, MoveNet, BlazePose, and PoseNet.

[0093] Specifically, the terminal device uses a lightweight human keypoint detection model to acquire keypoint data of the target person from real-time images. This keypoint data includes the positional data of all keypoints of the target person, such as the coordinates of the keypoints of the target person's limbs in the real-time image.

[0094] Step 102: Determine whether to push notification information by analyzing the key data of the target person.

[0095] Specifically, the terminal device analyzes the keypoint data of the target person to obtain at least one of the following: the number of keypoints of the target person, the corresponding detection box of the target person, and the target person's posture data. Based on the number of keypoints of the target person, the corresponding detection box of the target person, and at least one of the various posture data of the target person, it determines whether to push a prompt message. The prompt message is used to prompt the target person to adjust their standing position and / or posture.

[0096] In one optional embodiment of this example, if the terminal device determines to push notification information, the notification information is displayed on the camera interface of the terminal device, or the notification information is broadcast through the speaker of the terminal device.

[0097] In this embodiment, the terminal device analyzes the key point data of the target person to identify whether there is a person in the real-time image, whether the person's entire body is within the designated area of ​​the real-time image (i.e., whether the person's position meets the requirements), and whether the person's posture meets the requirements. If any of these requirements are not met, the terminal device pushes a prompt message. The purpose of the prompt message is to assist and guide the user in taking images that meet the preset posture requirements for subsequent calculation of anthropometric data.

[0098] The following detailed explanation, in conjunction with the accompanying drawings, illustrates the process of the terminal device recognizing real-time images and providing information prompts.

[0099] In one optional implementation, the terminal device obtains the number of key points of the target person based on the key point data of the target person, and determines whether to push the first prompt information by comparing the number of key points of the target person with the preset key point threshold.

[0100] The preset keypoint threshold is the number of keypoints to be identified on the entire body of a person. The preset keypoint threshold varies depending on the algorithm settings; it also varies depending on the image category, i.e., the preset keypoint threshold is different for frontal and side views of a person.

[0101] Specifically, if the number of key points on the target person is less than a preset key point threshold, a first instruction message is pushed. This first instruction message instructs the target person to stand in a designated area of ​​the terminal device's shooting interface; that is, it instructs the target person to adjust their standing position.

[0102] Figure 4 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 1 .like Figure 4As shown, the real-time image displayed on the terminal device's shooting interface only contains facial key points, such as the eyes, ears, and nose. Based on this key point data, the terminal device determines the number of key points to be 5. Since this number is less than a preset key point threshold (insufficient key points), the terminal device can provide a text prompt on the shooting interface, such as... Figure 4 The message "Please stand in the designated area" is displayed. The terminal device can also provide voice prompts via speaker, or both text and voice prompts simultaneously.

[0103] In one optional implementation, the terminal device determines the detection box corresponding to the target person based on the key point data of the target person; obtains the area ratio of the detection box to the real-time image, and determines whether to push a second prompt message by comparing the size relationship between the area ratio and a preset area ratio threshold.

[0104] Specifically, if the area ratio of the detection box to the real-time image is less than a preset area ratio threshold, a second prompt message is pushed. This second prompt message instructs the target person to move forward to a designated area on the shooting interface; that is, it instructs the target person to adjust their standing position.

[0105] Figure 5 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 2 .like Figure 5 As shown, the real-time image of the terminal device's shooting interface contains a preset number of key points representing the person. Based on these key points, a detection box for the person in the real-time image is determined. The area ratio of the person's detection box to the area of ​​the real-time image is obtained. Since this area ratio is less than a preset area ratio threshold, meaning the distance between the person and the terminal device is greater than a preset distance value, the terminal device can display a text prompt on the shooting interface, such as... Figure 5 The message displayed is "Please move forward to the designated area." The terminal device can also provide voice prompts via speaker, or both text and voice prompts simultaneously.

[0106] In one optional implementation, the terminal device determines the target person's posture data based on the target person's key point data; determines whether the target person's posture data is within the preset value range corresponding to the posture data, and thus determines whether to push a third prompt message.

[0107] Specifically, if the target person's posture data is outside the preset range corresponding to the posture data, a third prompt message will be pushed. The third prompt message is used to instruct the target person to adjust their standing posture.

[0108] Optionally, the target person's posture data may include upper limb posture data, lower limb posture data, head posture data, shoulder posture data, and torso posture data.

[0109] It should be noted that the posture data requirements for each part of the human body (i.e., the preset value range corresponding to the posture data of each part) can be set according to actual needs, and this embodiment does not impose specific limitations on this.

[0110] For example, if the target person's upper limb posture data is outside the preset value range corresponding to the upper limb posture data, a third prompt message will be pushed.

[0111] For example, if the target person's lower limb posture data is outside the preset value range corresponding to the lower limb posture data, a third prompt message will be pushed.

[0112] For example, if the target person's head posture data is outside the preset value range corresponding to the head posture data, a third prompt message will be pushed.

[0113] For example, if the target person's shoulder posture data is outside the preset value range corresponding to the shoulder posture data, a third prompt message will be pushed.

[0114] For example, if the target person's torso posture data is outside the preset value range corresponding to the torso posture data, a third prompt message will be pushed.

[0115] Optionally, the target person's posture data includes frontal posture data and side posture data. Both frontal and side posture data include the aforementioned posture data types.

[0116] Figure 6 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 3 .like Figure 6 As shown, the real-time image captured by the terminal device contains a preset number of key points on the person, and the area ratio of the person detection box to the real-time image reaches a preset area ratio threshold. However, the person is standing face-up with their arms not extended. Based on the key point data of the person's upper limbs, the terminal device obtains the person's upper limb posture data. If the upper limb posture data is determined to be outside the preset value range corresponding to the upper limb posture data, the terminal device can provide text prompts on the capturing interface, such as... Figure 6 The message displayed is "Please extend your arms at a 45-degree angle." The terminal device can also provide voice prompts via speaker, or simultaneously provide text and voice prompts.

[0117] It should be noted that, Figure 6 The illustrated embodiment is only one example of posture adjustment. Other actions that do not meet the preset posture requirements, such as bent legs or uneven shoulders, are all within the scope of posture adjustment in the embodiments of this application.

[0118] Step 103: Obtain a full-body image of the target person whose standing position and posture meet the preset requirements.

[0119] In this embodiment, the full-body image includes a frontal full-body image and a side full-body image of the target person. Acquiring a full-body image of the target person whose standing position and posture both meet preset requirements includes: acquiring a frontal full-body image of the target person whose standing position and posture both meet preset frontal pose requirements, and a side full-body image of the target person whose standing position and posture both meet preset side pose requirements.

[0120] For example, the preset frontal posture requirements include standing upright with feet shoulder-width apart and arms raised at a 45-degree angle. The preset side posture requirements include standing upright from the side with arms close to the body.

[0121] Step 104: Determine the anthropometric data of the target person based on the key point data of the target person in the full-body image. The anthropometric data is used to indicate the measurement data of various parts of the target person's body.

[0122] After acquiring a full-body image (front view) and a full-body image (side view) of the target person whose standing position and posture meet preset requirements, the terminal device uses a lightweight human keypoint detection model to obtain the first keypoint data of the target person from the full-body image and the second keypoint data from the full-body image. Based on the first and second keypoint data, the device determines the target person's anthropometric data, including body measurements such as waist, hip, shoulder width, neck circumference, arm length, and leg length.

[0123] Optionally, in some embodiments, after determining the anthropometric data of the target person, the terminal device stores the anthropometric data of the target person in the terminal device.

[0124] In this embodiment, by acquiring full-body front and side images of the user that meet the standing posture requirements at the terminal side, the key point data of the human body in the images is obtained. The measurement data of each part of the user's body is calculated based on the key point data. The measurement data is not uploaded to the server, but is stored at the terminal side to ensure the security of the user's privacy data.

[0125] The method for acquiring anthropometric data illustrated in this application involves a terminal device acquiring key point data of a target person in a real-time image from a shooting interface. Based on this key point data, the device determines whether to push a prompt message to the target person, prompting them to adjust their standing position and / or posture. After the target person adjusts their standing position and posture according to the prompt message, the terminal device acquires a full-body image of the target person whose standing position and posture both meet preset requirements. The device then determines the target person's anthropometric data based on the key point data in the full-body image. This process involves posture recognition of the person's key point data. By recognizing the posture, the device determines whether a shooting prompt is needed during the shooting process, thereby guiding the target person to acquire full-body front and side images, achieving rapid acquisition of images that meet the standing posture requirements, and improving the accuracy of the anthropometric data.

[0126] Figure 7 A schematic diagram of the shooting interface of the terminal device provided in the embodiments of this application. Figure 4 .exist Figure 4 , Figure 5 or Figure 6 Based on the shooting interface shown, as Figure 7 As shown, a standard pose display area can be added to the lower right corner of the shooting interface to display the standard pose in the current shooting scene, so as to help users adjust their pose.

[0127] Based on the above embodiments, optionally, the number of key points of the target person includes the number of key points of the target person's face and the number of key points of the target person's body.

[0128] Figure 8 Schematic diagram of key points in the full-body frontal image of a person provided in the embodiments of this application. Figure 1 .like Figure 8 As shown, the key points of the person's face include nose (0), left eye (1), right eye (2), left ear (3), and right ear (4). The key points of the person's body include left shoulder (5), right shoulder (6), left elbow (7), right elbow (8), left wrist (9), right wrist (10), left hip (11), right hip (12), left knee (13), right knee (14), left ankle (15), and right ankle (16). Therefore, the total number of key points in the full-body frontal image of the person is 17.

[0129] Optionally, preset keypoint thresholds include facial keypoint thresholds and body keypoint thresholds.

[0130] As an example, for a full-body frontal image of a person, the facial keypoint threshold is 1, the body keypoint threshold is 12, and the total preset keypoint threshold is 13. In one scenario, if the number of keypoints on the target person's face is 0, or the number of keypoints on the target person's body is less than 12, it indicates that the target person has not fully entered the designated area, and the terminal device pushes a first prompt message, indicating that the target person is standing in the designated area. In another scenario, if the total number of keypoints on the target person is less than 13, it indicates that the target person has not fully entered the designated area, and the terminal device pushes a first prompt message.

[0131] As an example, for a full-body side profile image of a person, the facial keypoint threshold is 1, the body keypoint threshold is 6, and the total preset keypoint threshold is 7. In one scenario, if the number of keypoints on the target person's face is 0, or the number of keypoints on the target person's body is less than 6, it may indicate that the target person has not fully entered the designated area, and the terminal device will push a first prompt message indicating that the target person is standing in the designated area. In another scenario, if the total number of keypoints on the target person is less than 7, it indicates that the target person has not fully entered the designated area, and the terminal device will push a first prompt message.

[0132] Human keypoint detection models can perform fine-grained detection on the human face, but instead detect the human head, for example, only the top of the head and the chin.

[0133] Optionally, the number of key points of the target person includes the number of key points on the head of the target person and the number of key points on the body of the target person.

[0134] Figure 9 Schematic diagram of key points in the full-body frontal image of a person provided in the embodiments of this application. Figure 2 .like Figure 9 As shown, the key points of the person's head include the top of the head (0) and the chin (1). The key points of the person's body include the left shoulder (2), right shoulder (3), left elbow (4), right elbow (5), left wrist (6), right wrist (7), left hip (8), right hip (9), left knee (10), right knee (11), left ankle (12), and right ankle (13). Therefore, the total number of key points in the full-body frontal image of the person is 14.

[0135] Optionally, preset keypoint thresholds include head keypoint thresholds and body keypoint thresholds.

[0136] As an example, for a full-body frontal image of a person, the threshold for head keypoints is 1, the threshold for body keypoints is 12, and the total preset keypoint threshold is 13. Specific details can be found in the examples above, and will not be elaborated upon here.

[0137] As an example, for a full-body side profile image of a person, the threshold for head keypoints is 1, the threshold for body keypoints is 6, and the total preset keypoint threshold is 7. Specific details can be found in the examples above, and will not be elaborated upon here.

[0138] Based on the above embodiments, optionally, the target person's posture data includes at least one of the following: upper limb opening and closing degree; shoulder offset degree; trunk offset degree; lower limb flexion degree; upper limb verticality.

[0139] Specifically, the upper limb opening and closing angle is determined based on the location data of the target person's upper limb key points; the shoulder offset is determined based on the location data of the target person's left and right shoulder key points; the torso offset is determined based on the location data of the target person's four limb key points and height. The lower limb flexion is determined based on the location data of the target person's lower limb key points and height; and the upper limb verticality is determined based on the location data of the target person's upper limb key points and hip key points.

[0140] The following is combined Figure 10 and Figure 11 The determination of the above-mentioned attitude data is explained in detail.

[0141] Figure 10 This is a schematic diagram of the posture data of a full-body frontal image of a person provided in an embodiment of this application.

[0142] For example, the upper limb opening and closing range is based on Figure 10 The positional data of key points 5-10 on the upper limbs of the figure shown were determined. Specifically, the data was obtained... Figure 10 The average angle between the four lines shown (a1, a2, a3, a4) and the central axis O1 of the frontal figure is calculated. If the average value is outside the range of 40 to 50 degrees, it indicates that the figure's arms are not extended at a 45-degree angle, and the terminal device will push a corresponding posture adjustment prompt. The central axis of the frontal figure can be determined by combining multiple pairs of symmetrical points on both sides of the central axis.

[0143] For example, the shoulder offset is based on Figure 10 The positions of key points 5 and 6 on the left and right shoulders of the figure shown were determined by data from the data. Specifically, the distance difference between the left and right shoulders and the central axis O1 of the figure from the front view was obtained, i.e., based on... Figure 10 The distance difference between d3 and d4 is shown, which represents the offset of the left and right shoulders of the person. If the absolute value of the distance difference between d3 and d4 is greater than the preset distance value (e.g., 10cm), and the left and right shoulders are located on both sides of the central axis O1 of the person in front, it indicates that the person is not standing facing forward and has an angular offset. The terminal device will push a prompt message to adjust the posture accordingly.

[0144] For example, the torso offset is based on Figure 10The location data of key points 5, 11, and 13 of the limbs of the shown figure, as well as the figure's height, are used to determine this. Specifically, the distance d1 from the left hip 11 to the line connecting the left shoulder 5 and the left knee 13 is obtained. The ratio of d1 to the figure's height is used as the torso offset, representing the curvature of the figure's upper body. If the ratio of d1 to the figure's height is greater than a preset ratio (e.g., 7%), it indicates that the figure's upper body is not upright, and the terminal device pushes a corresponding posture adjustment prompt. The figure's height is basic data pre-input by the user, and the basic data also includes the figure's weight.

[0145] For example, the lower limb curvature is based on Figure 10 The location data of key points 11, 13, and 15 of the lower limbs of the shown figure, as well as the figure's height, are determined. Specifically, the distance d2 from the left knee 13 to the line connecting the left hip 11 and the left ankle 15 is obtained, and the ratio of d2 to the figure's height is used as the lower limb flexion. If the ratio of d2 to the figure's height is greater than a preset ratio (e.g., 7%), it indicates that the figure's lower body is not upright, and the terminal device pushes a corresponding posture adjustment prompt.

[0146] Figure 11 This is a schematic diagram of the posture data of a full-body side view image of a person provided in an embodiment of this application.

[0147] For example, upper limb verticality is based on Figure 11 The positional data of key points 0 and 1 on the upper limbs and key point 2 on the hip of the person shown were determined. Specifically, the angle between line b1 connecting the left wrist 1 and the left shoulder 0 and line b2 connecting the left shoulder 0 and the hip 2 was obtained. The angle value can represent the verticality of the person's left arm. If the angle value is greater than a preset angle value (e.g., 30°), it indicates that the person's left arm is not vertically close to the body, and the terminal device pushes a corresponding posture adjustment prompt.

[0148] For example, the shoulder offset is based on Figure 1 The positions of the key points 0 and 0' on the left and right shoulders of the person shown are determined by the data. If there is an angular deviation in the shoulders when the person is standing sideways, the left shoulder 0 and right shoulder 0' will be detected in the full-body side view image. It can be understood that the left and right shoulder points coincide in a standard side-standing posture. If the left and right shoulders are located on one side of the side-standing central axis O2 of the person, and the average distance between the left and right shoulders and the side-standing central axis O2 ((c1+c2) / 2) is greater than a preset distance value (e.g., 10cm), it indicates that the person is not standing directly sideways and there is an angular deviation. The terminal device will push a corresponding posture adjustment prompt. The side-standing central axis of the person can be determined by the line connecting the top of the person's head and the bottom of their feet.

[0149] It should be noted that the above-mentioned attitude data are only examples. Other types of attitude data can be designed based on the same or similar principles, which will not be shown in this embodiment.

[0150] Figure 12This is a flowchart illustrating a processing method based on anthropometric data provided in an embodiment of this application. The processing method provided in this embodiment is applied to a terminal device, where the device's item interface displays the target item currently being viewed by the target person. For example, the item interface is the details page of an item in an e-commerce application app. Figure 12 As shown, the method includes the following steps:

[0151] Step 201: When a trigger operation on the processing control of the item interface is detected, obtain the standard size data of the target item from the server.

[0152] Optionally, the processing control can be an add-to-shop control or a size lookup control.

[0153] In this embodiment, the target item can be clothing items such as tops, pants, and hats. This embodiment does not specifically limit this item, and all items related to human body measurement data are within the protection scope of this embodiment.

[0154] Step 202: If the terminal device does not include the anthropometric data of the target person, control the entry into the shooting interface, and obtain a full-body image of the target person whose standing position and posture meet the preset requirements in the shooting interface.

[0155] Step 203: Determine the anthropometric data of the target person by acquiring key point data of the target person in the full-body image.

[0156] Step 204: Determine the recommended size based on the target person's anthropometric data and standard size data.

[0157] After the terminal device enters the shooting interface, it displays a real-time image and acquires key point data of the target person in the real-time image. The key point data is analyzed to determine whether to push a notification message, which prompts the target person to adjust their standing position and / or posture. The method for acquiring the key point data of the target person is the same as in the previous embodiment and will not be repeated here.

[0158] As an example, if the target person's standing position and / or posture do not meet the preset requirements, a prompt message will be pushed to the shooting interface.

[0159] As an example, if the target person's standing position and / or posture do not meet the preset requirements, a prompt message will be broadcast through the speaker of the terminal device.

[0160] As an example, if the target person's standing position and / or posture do not meet the preset requirements, a prompt message will be pushed to the shooting interface, and a prompt message will be broadcast through the speaker of the terminal device.

[0161] After the target person adjusts their standing position and posture according to the prompts on the shooting interface and / or the voice prompts, the terminal device acquires a full-body image of the target person whose standing position and posture meet the preset requirements. Based on a lightweight human keypoint detection model, keypoint data of the target person is obtained from the full-body image. Human body measurement data of the target person is determined based on the keypoint data in the full-body image. This human body measurement data is used to indicate the measurement data of various parts of the target person's body. The full-body image includes a full-body frontal image and a full-body side image. The above process can be referred to the embodiment above and will not be repeated here.

[0162] Optionally, in some embodiments, if the terminal device does not include the target person's anthropometric data, it controls the user to enter the information acquisition interface before controlling the user to enter the shooting interface. After the target person enters basic data including height and weight on the information acquisition interface, the user enters the shooting interface.

[0163] Optionally, in some embodiments, if the terminal device does not include the anthropometric data of the target person, the terminal device controls the entry into the shooting interface. After acquiring a full-body image of the target person that meets the requirements, it controls the entry into the information acquisition interface. The target person inputs the aforementioned basic data in the information acquisition interface, and then the terminal device controls the entry into the body measurement display interface.

[0164] Optionally, in some embodiments, if the terminal device includes the anthropometric data of the target person, the recommended size of the target item is determined directly based on the anthropometric data of the target person and the standard size data of the target item. In this embodiment, the anthropometric data of the target person is historical anthropometric data stored on the terminal device after the target person has taken their measurements in advance. The terminal device can directly obtain the anthropometric data of the target person from the local storage based on the target person's current operation, and after determining the recommended size, display the recommended size or add the item to the shopping cart.

[0165] As an example, if the control is for adding items, the terminal device, after determining the recommended size, generates the add-on item information for the target person based on the recommended size and the identifier of the target item.

[0166] As an example, if the processing control is a size lookup control, after the terminal device determines the recommended size, it will enter the body measurement display interface, which will display the target person's body measurement data and the recommended size of the target item.

[0167] Figure 13 This is a schematic diagram illustrating the changes in the graphical user interface of the terminal device provided in the embodiments of this application. For example... Figure 13As shown, the graphical user interface 301 includes information such as images, videos, prices, and feature descriptions of the target item. It also includes controls 302 and 303, where control 302 is used to trigger a size lookup and control 303 is used to add the target item to the shopping cart.

[0168] As an example, when a user triggers control 302, if the terminal device has the user's anthropometric data stored in it, the interface will jump to the body measurement display interface 304, which will display the appropriate size of the target item recommended to the user.

[0169] As an example, if the user triggers control 302 and the terminal device does not have the user's anthropometric data, the interface will jump to the shooting interface 305. By obtaining a full-body image of the front and side views that meets the standing posture requirements in the shooting interface 305, the user's anthropometric data will be determined, and then the user will jump to the body measurement display interface 304.

[0170] Optionally, if the terminal device does not have pre-stored basic data such as the user's height and weight, the control interface can jump to the information acquisition interface (not shown) before or after taking the front and side images to obtain the user's basic data.

[0171] As an example, if the terminal device has the user's anthropometric data stored in it, the user can directly add the target item of the recommended size to the shopping cart after the recommended size is determined.

[0172] As an example, if the user triggers control 303 and the terminal device does not have the user's anthropometric data, the interface jumps to the shooting interface 305. By acquiring a full-body image from the front and side views that meets the standing posture requirements in the shooting interface 305, the user's anthropometric data is determined, and thus the recommended size is determined. When the user returns to the item interface, the recommended size item is directly added to the user's shopping cart.

[0173] The anthropometric data processing method provided in this embodiment can meet users' needs to quickly obtain recommended sizes or add items to their cart while browsing products. If the terminal does not have records of the user's anthropometric data, a camera can be quickly started on the product browsing interface. Through data extraction and analysis by the terminal's lightweight human key point detection model, the user's anthropometric data can be quickly obtained, and product size recommendations or additions to the cart can be completed.

[0174] The foregoing described the method for acquiring human body measurement data and the method for processing human body measurement data provided in the embodiments of this application. The following will describe the device for acquiring human body measurement data and the device for processing human body measurement data provided in the embodiments of this application.

[0175] This application embodiment can divide the human body measurement data acquisition device and the human body measurement data processing device into functional modules according to the above method embodiment. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or software functional modules. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods. The following description uses the division of functional modules according to each function as an example.

[0176] Figure 14 Schematic diagram of the structure of the human body measurement data acquisition device provided in the embodiments of this application Figure 1 .like Figure 14 As shown, the human body measurement data acquisition device 400 provided in this embodiment includes: an acquisition module 401 and a processing module 402.

[0177] The acquisition module 401 is used to acquire key point data of the target person in the real-time image of the terminal device's shooting interface;

[0178] Processing module 402 is used to determine whether to push a prompt message by analyzing the key point data of the target person, the prompt message being used to prompt the target person to adjust their standing position and / or posture;

[0179] The acquisition module 401 is also used to acquire a full-body image of the target person whose standing position and posture both meet the preset requirements;

[0180] The processing module 402 is further configured to determine the human body measurement data of the target person based on the key point data of the target person in the full-body image, wherein the human body measurement data is used to indicate the measurement data of each part of the target person's body.

[0181] In an optional embodiment of this example, the acquisition module 401 is configured to:

[0182] Based on a lightweight human keypoint detection model, keypoint data of the target person is obtained from the real-time image. The keypoint data includes the position data of all keypoints of the target person.

[0183] In an optional embodiment of the first aspect of this application, the processing module 402 is configured to:

[0184] The number of key points is obtained based on the key point data of the target person;

[0185] If the number of key points is less than a preset key point threshold, the first instruction information is pushed. The first instruction information is used to instruct the target person to stand in a designated area of ​​the shooting interface.

[0186] In an optional embodiment of this example, the processing module 402 is configured to:

[0187] Determine the detection box corresponding to the target person based on the key point data of the target person;

[0188] Obtain the area ratio of the detection box to the real-time image;

[0189] If the area ratio is less than a preset area ratio threshold, a second prompt message is pushed. The second prompt message is used to instruct the target person to move forward to a designated area of ​​the shooting interface.

[0190] In an optional embodiment of this example, the processing module 402 is configured to:

[0191] The posture data of the target person is determined based on the key point data of the target person;

[0192] If the posture data is outside the preset value range corresponding to the posture data, a third prompt message is pushed. The third prompt message is used to instruct the target person to adjust their standing posture.

[0193] In an optional embodiment of this example, the attitude data includes at least one of the following:

[0194] Upper limb opening and closing degree, which is determined based on the position data of key points of the upper limbs of the target person;

[0195] Shoulder offset, which is determined based on the position data of the key points of the left and right shoulders of the target person;

[0196] The torso offset is determined based on the position data of the key points of the target person's limbs and height.

[0197] Lower limb curvature, which is determined based on the location data of key points of the target person's lower limbs and height;

[0198] Upper limb verticality, which is determined based on the positional data of the key points of the upper limbs and the key points of the buttocks of the target person.

[0199] Figure 15 Schematic diagram of the structure of the human body measurement data acquisition device provided in the embodiments of this application Figure 2 .exist Figure 14 Based on the device shown, such as Figure 15As shown, the human body measurement data acquisition device 400 provided in this embodiment also includes: a display module 403, a voice broadcast module 404, and a storage module 405.

[0200] In an optional embodiment of this example, if the processing module 402 determines to push the prompt information, the display module 403 is used to display the prompt information on the shooting interface of the terminal device, or the voice broadcasting module 404 is used to broadcast the prompt information through the speaker of the terminal device.

[0201] In an optional embodiment of this example, the full-body image includes a frontal full-body image and a side full-body image of the target person; the acquisition module 401 is used to acquire first key point data of the target person in the frontal full-body image and second key point data of the target person in the side full-body image.

[0202] The processing module 402 is used to determine the anthropometric data of the target person based on the first key point data and the second key point data.

[0203] In an optional embodiment of this example, the storage module 405 is used to store the anthropometric data of the target person in the terminal device.

[0204] The anthropometric data acquisition device provided in this embodiment can perform... Figure 3 The technical solutions of the method embodiments shown are similar in principle and in effect, and will not be described again here.

[0205] Figure 16 This is a schematic diagram of the structure of a processing device based on human body measurement data provided in an embodiment of this application. Figure 16 As shown, the human body measurement data processing device 500 provided in this embodiment includes: an acquisition module 501, a processing module 502, and a display module 503.

[0206] The acquisition module 501 is used to acquire the standard size data of the target item from the server when a trigger operation is detected on the processing control of the item interface of the terminal device.

[0207] The processing module 502 is used to control the entry into the shooting interface if the terminal device does not include the anthropometric data of the target person. The acquisition module 501 is also used to acquire a full-body image of the target person whose standing position and posture meet the preset requirements in the shooting interface. The shooting interface includes prompt information, which is used to prompt the target person to adjust their standing position and / or posture.

[0208] The processing module 502 is further configured to determine the anthropometric data of the target person by acquiring key point data of the target person in the full-body image, and determine the recommended size based on the anthropometric data of the target person and the standard size data.

[0209] In an optional embodiment of this example, if the processing control is a control for adding items, the processing module 502 is further configured to generate add-on item information for the target person based on the recommended size and the identifier of the target item after determining the recommended size.

[0210] In an optional embodiment of this example, if the processing control is a size query control, the processing module 502 is further configured to control entry into the measurement display interface after determining the recommended size;

[0211] Display module 503 is used to display the anthropometric data of the target person and the recommended size of the target item on the body measurement display interface.

[0212] The human anthropometric data processing device provided in this embodiment can perform... Figure 12 The technical solutions of the method embodiments shown are similar in principle and in effect, and will not be described again here.

[0213] Figure 17 This is a hardware structure diagram of the electronic device provided in an embodiment of this application. For example... Figure 17 As shown, the electronic device 600 provided in this embodiment includes: a memory 601, a processor 602, and a computer program. The computer program is stored in the memory 601 and configured to be executed by the processor 602 to implement the technical solutions of any of the above method embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0214] Optionally, the memory 601 can be either standalone or integrated with the processor 602. When the memory 601 is a device independent of the processor 602, the electronic device 600 also includes a bus 603 for connecting the memory 601 and the processor 602.

[0215] This application also provides a computer-readable storage medium storing a computer program thereon, which is executed by a processor 602 to implement the technical solutions of any of the foregoing method embodiments.

[0216] This application provides a computer program product, including a computer program, which, when executed by a processor, implements the technical solutions of any of the foregoing method embodiments.

[0217] This application also provides a chip, including: a processing module and a communication interface, wherein the processing module is capable of executing the technical solutions of any of the foregoing method embodiments.

[0218] Furthermore, the chip also includes a storage module (e.g., a memory), which is used to store instructions, and a processing module is used to execute the instructions stored in the storage module. The execution of the instructions stored in the storage module causes the processing module to execute the technical solution of any of the aforementioned method embodiments.

[0219] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0220] The memory may include high-speed RAM, and may also include non-volatile storage (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0221] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0222] The aforementioned storage medium can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage medium can be any available medium accessible to general-purpose or special-purpose computers.

[0223] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. Both the processor and the storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic device.

[0224] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for acquiring anthropometric data, characterized in that, The method is applied to a terminal device, wherein the terminal device's camera interface displays a real-time image; the method includes: Acquire key point data of the target person in the real-time image; By analyzing the key point data of the target person, it is determined whether to push a prompt message. The prompt message is used to prompt the target person to adjust their standing position and / or posture. A full-body image of the target person whose standing position and posture meet the preset requirements is obtained, and the anthropometric data of the target person is determined based on the key point data of the target person in the full-body image. The anthropometric data is used to indicate the measurement data of each part of the target person's body. The step of determining whether to push a notification message by analyzing the key data of the target person includes: The posture data of the target person is determined based on the key point data of the target person; wherein, the posture data includes at least two of the following: upper limb opening and closing degree, shoulder offset degree, trunk offset degree, lower limb flexion degree, and upper limb verticality; the shoulder offset degree is determined based on the position data of the left and right shoulder key points of the target person; the trunk offset degree is the ratio of a first distance value to the height of the target person, the first distance value being the vertical distance from the left hip key point of the target person to a first line, the first line being the line connecting the left shoulder key point and the left knee key point of the target person; the lower limb flexion degree is the ratio of a second distance value to the height of the target person, and so on. The second distance value is the vertical distance from the target person's left knee key point to the second connecting line, where the second connecting line is the line connecting the target person's left hip key point and the target person's left ankle key point; the upper limb verticality is the angle between the target person's third connecting line and the fourth connecting line, where the third connecting line is the line connecting the target person's left wrist key point and the target person's left shoulder key point, and the fourth connecting line is the line connecting the target person's left shoulder key point and the target person's hip key point; if the posture data is outside the preset value range corresponding to the posture data, a third prompt message is pushed, which is used to instruct the target person to adjust their standing posture.

2. The method according to claim 1, characterized in that, The acquisition of key point data of the target person in the real-time image includes: Based on a lightweight human keypoint detection model, keypoint data of the target person is obtained from the real-time image. The keypoint data includes the position data of all keypoints of the target person.

3. The method according to claim 1, characterized in that, The step of determining whether to push a notification message by analyzing the key point data of the target person also includes: The number of key points is obtained based on the key point data of the target person; If the number of key points is less than a preset key point threshold, a first instruction message is pushed. The first instruction message is used to instruct the target person to stand in a designated area of ​​the shooting interface.

4. The method according to claim 1 or 3, characterized in that, The step of determining whether to push a notification message by analyzing the key point data of the target person also includes: Determine the detection box corresponding to the target person based on the key point data of the target person; Obtain the area ratio of the detection box to the real-time image; If the area ratio is less than a preset area ratio threshold, a second prompt message is pushed. The second prompt message is used to instruct the target person to move forward to a designated area of ​​the shooting interface.

5. The method according to claim 1, characterized in that, in, The opening and closing degree of the upper limbs is determined based on the positional data of key points of the target person's upper limbs.

6. The method according to any one of claims 1-2 and 5, characterized in that, The method further includes: If the notification message is to be pushed, the notification message will be displayed on the camera interface of the terminal device, or the notification message will be broadcast through the speaker of the terminal device.

7. The method according to any one of claims 1-2 and 5, characterized in that, The full-body image includes a frontal full-body image and a side full-body image of the target person; determining the anthropometric data of the target person based on the key point data of the target person in the full-body image includes: Acquire the first key point data of the target person in the frontal full-body image and the second key point data of the target person in the side full-body image; Based on the first key point data and the second key point data, the anthropometric data of the target person are determined.

8. A method for processing anthropometric data, characterized in that, Applied to a terminal device, the terminal device's item interface displays the target item currently being viewed by the target person; the method includes: When a trigger operation on the processing control of the item interface is detected, the standard size data of the target item is obtained from the server; If the terminal device does not include the anthropometric data of the target person, the system will enter the shooting interface and acquire a full-body image of the target person whose standing position and posture meet the preset requirements. The shooting interface includes prompt information, which is used to prompt the target person to adjust their standing position and / or posture. By acquiring key point data of the target person in the full-body image, the anthropometric data of the target person is determined, and a recommended size is determined based on the anthropometric data of the target person and the standard size data. The full-body image is obtained by the method described in any one of claims 1-7.

9. The method according to claim 8, characterized in that, If the processing control is a control for adding items to a shopping cart, the method further includes: After determining the recommended size, add-on item information for the target person is generated based on the recommended size and the identifier of the target item.

10. The method according to claim 8, characterized in that, If the processing control is a size lookup control, the method further includes: After determining the recommended size, the system enters the body measurement display interface, which displays the anthropometric data of the target person and the recommended size of the target item.

11. An electronic device, characterized in that, include: A memory, a processor, and a computer program; the computer program is stored in the memory and configured to be executed by the processor to implement the method as claimed in any one of claims 1-10.

Images