Method for measuring human body data by integrating visual sensor on mobile terminal
By integrating visual sensors into mobile terminals, the system automatically generates 3D models, solving the data accuracy problem when purchasing clothing online. This achieves a precise wearing simulation effect and improves the user experience.
Patent Information
- Application Number
- CN202511576005.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2025-05-12
- Filing Date
- 2025-10-31
- Publication Date
- 2026-02-10
AI Technical Summary
When purchasing clothing online, users cannot provide accurate 3D data for a fitting simulation, resulting in a poor purchasing experience.
By integrating visual sensors into mobile terminals, the system automatically scans and generates 3D models, including hardware detection, data acquisition, processing, and modeling, producing high-precision 3D human body models, which are then uploaded to e-commerce platforms or virtual fitting devices.
It enables users to measure their own 3D data and generate accurate 3D models for online simulation of wearability, thus enhancing the purchasing experience.
Smart Images
Figure CN121504564A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of human body data measurement technology, and in particular to a method for measuring human body data using a mobile terminal with an integrated visual sensor. Background Technology
[0002] Currently, when people buy clothes online, they cannot try them on and can only judge the effect by looking at the models wearing the products. However, everyone's height, weight, bust, waist, and hip measurements are different, which leads to the difference between the effect on different users and the effect shown on the models after purchase. This results in problems such as returns and exchanges, which affects the customer's shopping experience.
[0003] Currently, some online platforms require customers to manually input their height, weight, bust, waist, and hip measurements to recommend suitable clothing. However, the data input by customers is only partial and cannot accurately generate corresponding 3D models for wearing simulation. Consequently, it is also impossible to effectively demonstrate the simulated effect of wearing the clothes to customers. Therefore, there is a need for a method that allows customers to easily provide accurate 3D data themselves, thereby enabling wearing simulation based on the 3D data and providing a better purchasing experience. Summary of the Invention
[0004] The purpose of this invention is to provide a method for measuring human body data by integrating a visual sensor into a mobile terminal. This allows customers to easily scan their own 3D data using a mobile terminal, thereby generating a corresponding 3D model for wearable simulation.
[0005] The technical solution adopted in the method for measuring human body data by integrating a visual sensor in a mobile terminal disclosed in this invention is as follows: A method for measuring human body data using a mobile terminal with an integrated visual sensor includes the following steps; Hardware and environmental testing involves testing the built-in hardware of the mobile terminal to determine whether it is equipped with the necessary visual sensors, and testing whether the lighting conditions in the measurement environment meet the measurement requirements. The measurement guide informs the subject of the clothing requirements during the measurement, and provides information on the subject's location and range of motion during the measurement process. Data acquisition includes dynamic and static modes. The mobile terminal sensor information is obtained through the above process, and the system automatically switches to the corresponding acquisition mode according to the sensor type in the information. In dynamic mode, the subject needs to rotate to capture full-body data. In static mode, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition. Data processing and modeling, noise reduction and filtering: removing sensor noise or outliers through methods such as Gaussian filtering and statistical filtering; model reconstruction: human body modeling based on visual sensors utilizes color / texture information or depth data, combined with parametric human body models or 3D reconstruction techniques, to generate high-precision meshes and optimize physical consistency; Data display and data upload: The established 3D model is displayed through a mobile terminal, and the circumference data of each key part is displayed synchronously; after the subject confirms that it is correct, the data is uploaded.
[0006] As a preferred option, the visual sensor includes one or more combinations of RGB sensors and TOF sensors. After hardware testing, if the mobile terminal integrates an RGB sensor, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition; if the mobile terminal integrates a TOF sensor, the data acquisition modes include pulse wave modulation, continuous wave modulation, multi-point scanning, and dynamic timing.
[0007] As a preferred solution, the hardware detection methods for mobile terminals include: Android systems calling the Sensor API for hardware detection; iOS systems calling ARKit for hardware detection; and the lighting conditions are as follows: when the ToF sensor is iToF, if the ambient light is >5000 lux, the software prompts the user to change the measurement environment; when it is dToF, if the ambient light is >20000 lux, the user is prompted to change the measurement environment.
[0008] As a preferred option, the subject should wear a non-reflective bodysuit, and the subject's position should be monitored in real time during the measurement process to ensure that the subject is within the shooting range. If the subject goes out of range or moves too fast during the measurement process, a warning will be issued.
[0009] As a preferred method, the human body circumference measurement process is as follows: Locating key points: Using AI algorithms to detect key points for human body circumference measurement such as acromion, anterior superior iliac spine, chest height, and greater trochanter of waistline; 3D model slicing: On the 3D model, cut a horizontal cross-section along the vertical direction for each key point; Calculate the perimeter: Calculate the perimeter of the cross section at each key point; Correcting clothing errors: The total length is calculated by integrating along the edge contour line of the cut section, and the influence of clothing thickness is automatically deducted.
[0010] As a preferred option, the model creation process also includes texture mapping: by fusing TOF depth maps with color photos taken by a mobile phone, skin texture details are added to the model, making the virtual human body present realistic skin color and wrinkles.
[0011] As a preferred option, Laplace noise is used and an encryption algorithm is employed for encrypted data upload.
[0012] The beneficial effects of the method for measuring human body data by integrating a visual sensor into a mobile terminal disclosed in this invention are as follows: By integrating active three-dimensional perception hardware into the mobile terminal, users can measure their own body and three-dimensional height data or those of others. After obtaining these data, users generate corresponding 3D models, and after confirmation, upload the corresponding 3D model data to e-commerce platforms or virtual fitting devices in offline stores to generate a three-dimensional human body model. Thus, when purchasing clothing, users can use their own three-dimensional human body model to simulate the upper body effect for accurate fitting of clothing. Attached Figure Description
[0013] Figure 1 This is a flowchart of the method for measuring human body data by integrating a visual sensor into a mobile terminal according to the present invention. Detailed Implementation
[0014] The present invention will be further described and illustrated below with reference to specific embodiments and the accompanying drawings: Please refer to Figure 1 A method for measuring human body data using a mobile terminal with an integrated visual sensor, comprising the following steps; Hardware and environmental detection involves detecting the built-in hardware of the mobile terminal to determine whether it is equipped with the necessary visual sensors, including one or more combinations of RGB sensors and TOF sensors.
[0015] In addition, it detects whether the light in the measurement environment meets the measurement conditions. The hardware detection methods of the mobile terminal include: Android system calls the Sensor API for hardware detection; iOS system calls ARKit for hardware detection; the light conditions are: when the ToF sensor is iToF, if the ambient light is >5000 lux, the software prompts the user to change the measurement environment; when it is dToF, if the ambient light is >20000 lux, the subject is prompted to change the measurement environment. The test subject is guided on their clothing requirements, location, and body movements during the measurement. They should wear non-reflective, form-fitting clothing, and preferably avoid silk or metal jewelry, as these can cause glare and affect the measurement data. During the measurement, the test subject's position is monitored in real-time to ensure they remain within the shooting range. If they are outside the range or move too quickly during the measurement, the mobile terminal may fail to effectively capture the subject, leading to data errors and requiring timely warnings.
[0016] Data acquisition includes dynamic and static modes. By acquiring sensor information from the mobile terminal as described above, the system automatically switches to the corresponding acquisition mode based on the sensor type in the information, or prompts the photographer to switch to the corresponding data acquisition mode. In dynamic mode, the subject needs to rotate to capture full-body data; in static mode, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition.
[0017] If the mobile terminal integrates an RGB sensor, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition; if the mobile terminal integrates a TOF sensor, the data acquisition modes include pulse wave modulation, continuous wave modulation, multi-point scanning, and dynamic timing.
[0018] The postures adopted by the subject in the static mode include T-Pose and A-Pose, which means that the subject stands with both feet together and arms raised horizontally or overhead.
[0019] The process of measuring body circumference after data collection: Locating key points: Using AI algorithms to detect key points for human body circumference measurement such as acromion, anterior superior iliac spine, chest height, and greater trochanter of waistline; 3D model slicing: On the 3D model, cut a horizontal cross-section along the vertical direction for each key point; Calculate the perimeter: Calculate the perimeter of the cross section at each key point; Correcting clothing errors: The total length is calculated by integrating along the edge contour line of the cut section, and the influence of clothing thickness is automatically deducted.
[0020] Data processing and modeling, point cloud denoising: The system automatically filters isolated points floating in the air, such as dust and hair tips, to retain continuous point clouds of the human body surface. For motion blur-induced shadow areas, such as when swinging arms or rotating, time series analysis is used to remove outliers. Model building: By calling the backend API, multiple discrete points are connected into a triangular mesh. Sharp edges are eliminated using a surface smoothing algorithm to generate a smooth human body model. The model is scaled proportionally according to the calibration disk size to ensure that the virtual human body is completely consistent with the actual size.
[0021] The model creation process also includes texture mapping: by fusing TOF depth maps with color photos taken with a mobile phone, skin texture details are added to the model, making the virtual human body present realistic skin tone and wrinkles.
[0022] Data display and uploading: The established 3D model will be displayed via mobile terminal, and the circumference data of key parts, such as hip circumference, chest circumference, shoulder width, and waist circumference, will be displayed simultaneously. After the subject confirms that the data is correct, Laplace noise and encryption algorithm will be used to encrypt and upload the data.
[0023] This invention provides a method for measuring human body data by integrating a visual sensor into a mobile terminal. By integrating active three-dimensional sensing hardware into the mobile terminal, users can measure their own body circumference, height, and other human body dimensions, or measure the height of others. After obtaining these measurements, a corresponding 3D model is generated. After confirmation, the corresponding 3D model data is uploaded to e-commerce platforms or virtual fitting devices in offline stores to generate a three-dimensional human body model. Thus, when purchasing clothing, users can use their own three-dimensional human body model to simulate the effect of wearing the garment, enabling precise fitting of clothing items.
[0024] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.
Claims
1. A method for measuring human body data using a mobile terminal integrated with a visual sensor, characterized in that, Includes the following steps; Hardware and environmental testing involves testing the built-in hardware of the mobile terminal to determine whether it is equipped with the necessary visual sensors, and testing whether the lighting conditions in the measurement environment meet the measurement requirements. The measurement guide informs the subject of the clothing requirements during the measurement, and provides information on the subject's location and range of motion during the measurement process. Data acquisition includes dynamic and static modes. The mobile terminal sensor information is obtained through the above process, and the system automatically switches to the corresponding acquisition mode according to the sensor type in the information. In dynamic mode, the subject needs to rotate to capture full-body data. In static mode, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition. Data processing and modeling, noise reduction and filtering: removing sensor noise or outliers through methods such as Gaussian filtering and statistical filtering; model reconstruction: human body modeling based on visual sensors utilizes color / texture information or depth data, combined with parametric human body models or 3D reconstruction techniques, to generate high-precision meshes and optimize physical consistency; Data display and data upload: The established 3D model is displayed through a mobile terminal, and the circumference data of each key part is displayed synchronously; after the subject confirms that it is correct, the data is uploaded.
2. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 1, characterized in that, Visual sensors include one or more combinations of RGB sensors and TOF sensors. After hardware testing, such as when a mobile terminal integrates an RGB sensor, the subject needs to pose in a specified manner to take one or more photos to complete the corresponding data acquisition; if the mobile terminal integrates a TOF sensor, the data acquisition modes include pulse wave modulation, continuous wave modulation, multi-point scanning, and dynamic timing.
3. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 2, characterized in that, The hardware detection methods for mobile terminals include: Android systems call the Sensor API for hardware detection; iOS systems call ARKit for hardware detection; the lighting conditions are as follows: when the ToF sensor is iToF, if the ambient light is >5000 lux, the software prompts the user to change the measurement environment; when it is dToF, if the ambient light is >20000 lux, the subject is prompted to change the measurement environment.
4. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 1, characterized in that, The subject must wear a non-reflective, tight-fitting garment. During the measurement process, the subject's position is monitored in real time to ensure they remain within the shooting range. If they go out of range or move too quickly during the measurement, a warning will be issued.
5. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 1, characterized in that, Human body circumference measurement process: Locating key points: Using AI algorithms to detect key points for human body circumference measurement such as acromion, anterior superior iliac spine, chest height, and greater trochanter of waistline; 3D model slicing: On the 3D model, cut a horizontal cross-section along the vertical direction for each key point; Calculate the perimeter: Calculate the perimeter of the cross section at each key point; Correcting clothing errors: The total length is calculated by integrating along the edge contour line of the cut section, and the influence of clothing thickness is automatically deducted.
6. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 1, characterized in that, The model creation process also includes texture mapping: by fusing TOF depth maps with color photos taken with a mobile phone, skin texture details are added to the model, making the virtual human body present realistic skin tone and wrinkles.
7. The method for measuring human body data using a mobile terminal integrated with a visual sensor as described in claim 1, characterized in that, When uploading data, Laplace noise is used and an encryption algorithm is employed for encrypted uploading.