Human body size measurement method, product information recommendation method, human body size measurement method, and human body size measurement device
By identifying and reconstructing the human body model in the image and measuring its size data, the problem of low measurement efficiency of human body size is solved and efficient size data acquisition is achieved.
Patent Information
- Application Number
- CN202411026886.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
AI Technical Summary
In the prior art, the measurement efficiency of human body size is low and there is a lack of effective solutions.
By identifying multiple images to be processed, the human body model of the human body to be measured is reconstructed, and the model is measured, and the size data of the part is obtained to determine the predicted size data.
It improves the detection efficiency of human size data and solves the problem of low size measurement efficiency.
Smart Images

Figure CN119090934A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the fields of large model technology and data processing, and in particular, to a method for measuring human body size, a method for recommending product information, a method for measuring human body size, and a device for measuring human body size. Background Art
[0002] In the related art, the measurement of the size data of the human body to be measured usually relies on on-site measurement by professionals or measurement using professional scanning equipment. However, the above methods have the technical problem of low efficiency in detecting the size data of the human body.
[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention
[0004] The embodiments of the present application provide a method for measuring the size of a human body, a method for recommending product information, a method for measuring the size of a human body, and a device for measuring the size of a human body, so as to at least solve the technical problem of the efficiency of measuring the size data of an object.
[0005] According to one aspect of an embodiment of the present application, a method for detecting size data of an object is provided. The method may include: identifying multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; reconstructing the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measuring at least one part model in the human body model to obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; and determining the size data of the part model as the predicted size data of the corresponding part.
[0006] According to another aspect of the embodiment of the present application, a method for recommending product information is also provided. The method can be applied to an e-commerce platform and can include: identifying multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; reconstructing the human body to be measured in the processed images to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measuring at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determining the size data of the part model as the predicted size data of the corresponding part; recommending product attribute information matching the predicted size data, wherein the clothing products that meet the product attribute information are to be worn by the corresponding part, and the clothing products come from a virtual clothing store deployed on the e-commerce platform.
[0007] According to another aspect of the embodiment of the present application, another method for recommending product information is also provided. The method can be applied to a client, and the client is used to deploy a virtual clothing store on an e-commerce platform, and can include: identifying multiple images to be processed, wherein the image content of the multiple images to be processed includes the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; reconstructing the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measuring at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured; determining the size data of the part model as the predicted size data of the corresponding part; recommending product attribute information matching the predicted size data to the e-commerce platform, wherein the clothing products that meet the product attribute information are to be worn by the corresponding part, and the clothing products come from the virtual clothing store.
[0008] According to one aspect of the embodiment of the present application, a method for measuring the size of a human body is also provided. The method may include: in response to an image input operation on an operation interface, displaying a plurality of images to be processed on the operation interface, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; in response to a reconstruction operation on the operation interface, displaying a human body model of the human body to be measured on the operation interface, wherein the human body model is obtained by reconstructing the human body to be measured in the image to be processed and is used to simulate the geometric contour of the human body to be measured; in response to a size measurement operation on the operation interface, displaying predicted size data of at least one part of the human body to be measured on the operation interface, wherein the predicted size data of the part is determined based on the size data of the corresponding part model in the human body model, and the part model is used to simulate the geometric contour of the corresponding part.
[0009] According to one aspect of the embodiment of the present application, a method for measuring the size of a human body is also provided. The method may include: displaying multiple images to be processed on a presentation screen of a virtual reality VR device or an augmented reality AR device, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; driving the VR device or AR device to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; driving the VR device or AR device to measure at least one part model in the human body model to obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determining the size data of the part model as the predicted size data of the corresponding part, and displaying the predicted size data of the part on the presentation screen of the VR device or AR device.
[0010] According to one aspect of the embodiment of the present application, an image processing system is also provided. The system may include: a client, used to upload multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; a server, used to reconstruct the human body to be measured in the images to be processed, and obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measure at least one part model in the human body model, and obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determine the size data of the part model as the predicted size data of the corresponding part.
[0011] According to one aspect of the embodiment of the present application, a human body size measurement device is also provided. The device may include: a first recognition unit, used to recognize multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; a first reconstruction unit, used to reconstruct the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; a first measurement unit, used to measure at least one part model in the human body model to obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; a first determination unit, used to determine the size data of the part model as the predicted size data of the corresponding part.
[0012] According to another aspect of the embodiments of the present application, a computer terminal is further provided, including: a memory storing an executable program; and a processor for running the program, wherein the method in each embodiment of the present application is executed when the program is running.
[0013] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in each embodiment of the present application.
[0014] According to another aspect of the embodiments of the present application, a computer program product is also provided, including a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present application is implemented.
[0015] According to another aspect of an embodiment of the present application, a computer program product is also provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present application is implemented.
[0016] According to another aspect of the embodiments of the present application, a computer program is also provided, and when the computer program is executed by a processor, the methods in the various embodiments of the present application are implemented.
[0017] In the embodiment of the present application, multiple images to be processed are identified, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the human body to be measured in the images to be processed is reconstructed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; at least one part model in the human body model is measured to obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as the predicted size data of the corresponding part. That is, in the embodiment of the present application, multiple images to be processed can be obtained, for example, a front human body image and a side human body image, and a human body model of the human body to be measured (which can be a human body geometric mesh) can be reconstructed using the multiple images to be processed, and the size data of the human body to be measured (for example, human body size information) can be obtained by measuring the human body model, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0018] It is easy to notice that the above general description and the following detailed description are only for the purpose of exemplifying and explaining the present application, and do not constitute a limitation of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0020] Figure 1 is a schematic diagram of an application scenario of a method for measuring human body size according to an embodiment of the present application;
[0021] Figure 2 is a flow chart of a method for measuring human body size according to an embodiment of the present application;
[0022] Figure 3 is a flow chart of a method for measuring size data of a biological object according to an embodiment of the present application;
[0023] Figure 4 is a flow chart of a method for recommending product information according to an embodiment of the present application;
[0024] Figure 5 is a flow chart of another method for recommending product information according to an embodiment of the present application;
[0025] Figure 6 is another flow chart of measuring the size of a human body according to an embodiment of the present application;
[0026] Figure 7 is a flow chart of measuring size data of another object according to an embodiment of the present application;
[0027] Figure 8 is a schematic diagram of predicted size data of a sub-object according to an embodiment of the present application;
[0028] Fig. 9 is a schematic diagram of an image processing system according to an embodiment of the present application;
[0029] Fig.10 is a schematic diagram of a human body size measurement framework based on a differentiable measuring device according to an embodiment of the present application;
[0030] Fig.11 is a schematic diagram of a data sample according to an embodiment of the present application;
[0031] Fig.12 It is a hardware structure block diagram of a computer terminal (or mobile device) for a method for measuring size data of an object according to an embodiment of the present application;
[0032] Fig.13 is a schematic diagram of a human body size measuring device according to an embodiment of the present application;
[0033] Fig.14 is a schematic diagram of a device for measuring size data of a biological object according to an embodiment of the present application;
[0034] Fig.15is a schematic diagram of a device for recommending product information according to an embodiment of the present application;
[0035] Fig.16 is a schematic diagram of another device for recommending product information according to an embodiment of the present application;
[0036] Fig.17 is a schematic diagram of another human body size measuring device according to an embodiment of the present application;
[0037] Fig.18 is a schematic diagram of another human body size measuring device according to an embodiment of the present application;
[0038] Fig.19 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0039] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.
[0040] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0041] The technical solution provided in this application is mainly implemented by large-scale model technology. The large model here refers to a deep learning model with large-scale model parameters, which can usually contain hundreds of millions, tens of billions, hundreds of billions, trillions or even more than ten trillion model parameters. The large model can also be called the foundation model / foundation model (Foundation Model l). The large model is pre-trained through large-scale unlabeled corpus to produce a pre-trained model with more than 100 million parameters. This model can adapt to a wide range of downstream tasks, and the model has good generalization ability, such as large-scale language model (Large Language Model l, referred to as LLM), multi-modal pre-training model (multi-modal pre-training model l), etc.
[0042] It should be noted that when the large model is actually applied, the pre-trained model can be fine-tuned through a small number of samples so that the large model can be applied to different tasks. For example, the large model can be widely used in natural language processing (NLP), computer vision, speech processing and other fields, and can be specifically applied to computer vision tasks such as visual question answering (VQA), image description (IC), image generation, etc. It can also be widely used in natural language processing tasks such as text-based sentiment classification, text summary generation, and machine translation. Therefore, the main application scenarios of the large model include but are not limited to digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc. In the embodiment of the present application, the data processing of the image to be processed in the human body size measurement scenario is taken as an example to explain:
[0043] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:
[0044] Body size, which can refer to the size of certain parts of the body that affect the selection of clothing sizes, for example, it can include shoulder width, arm length, chest circumference, waist circumference, hip circumference, leg length, etc.;
[0045] Human key point detection can be to use a detection model, take a person picture as input, and detect the key point positions of the human skeleton in the picture, where the key point positions can be pixel coordinates;
[0046] Dimensional fitting can be used to measure the width of the corresponding body parts based on the front picture, and the thickness from the side picture to fit the body circumference, such as chest / waist / hip / circumference, etc.
[0047] Three-dimensional (3D) human body reconstruction refers to the process of reconstructing the geometric outline of a person using one or more images of the same person as input, which can be represented by a mesh.
[0048] Differentiable module refers to the module that allows error gradient back propagation when training the model, which is the basis for model training.
[0049] A differentiable measurer may be a straight line length, geodesic distance, or cross-section perimeter commonly used in 3D software to measure two points on a mesh. The measurer may be differentiable, allowing gradient backpropagation, and may be integrated into model training.
[0050] According to an embodiment of the present application, a method for measuring the size of a human body is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0051] According to a method of an embodiment of the present application, a method for measuring the size of a human body is provided. As an optional implementation, the above-mentioned method for measuring the size of a human body may include but is not limited to being applied to Figure 1 The application scenario shown. Figure 1 is a schematic diagram of an application scenario of a method for measuring human body size according to an embodiment of the present application, such as Figure 1 As shown, in the application scenario, the terminal device 12 can, but is not limited to, communicate with the server 16 through the network 14, for example, it can be used to transmit images to be processed, size data, etc., and the server 16 can, but is not limited to, perform operations on the database 18, for example, write data operations or read data operations. The above-mentioned terminal device 12 can, but is not limited to, include a human-computer interaction screen, a processor and a memory. The above-mentioned human-computer interaction screen can, but is not limited to, be used to display input information, target response information, etc. on the terminal device 12. The above-mentioned processor can include, but is not limited to, being used to respond to the above-mentioned human-computer interaction operation, perform corresponding operations, or generate corresponding instructions, and send the generated instructions to the server 16. The above-mentioned memory is used to store relevant processing data, such as human body models, size data, etc.
[0052] As an optional manner, the following steps in the human body size measurement method can be executed on the server 16: step S102, identifying multiple images to be processed; step S104, reconstructing the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured; step S106, measuring at least one part model in the human body model to obtain size data of the part model; step S108, determining the size data of the part model as predicted size data of the corresponding part.
[0053] By adopting the above method, in the embodiment of the present application, multiple images to be processed can be obtained, for example, a front human body image and a side human body image, and a human body model (which can be a human body geometric mesh) of the human body to be measured is reconstructed using the multiple images to be processed. By measuring the human body model, the size data of the human body to be measured (for example, human body size information) is obtained, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body, and solving the technical problem of the efficiency of measuring the size data of the object.
[0054] Under the above operating environment, this application provides Figure 2 The measurements of the human body shown are as follows. Figure 2 FIG. 1 is a flow chart of a method for measuring the size of a human body according to an embodiment of the present application. Figure 2 As shown, the method may include the following steps:
[0055] Step S202: identifying a plurality of images to be processed.
[0056] In the technical solution provided in the above step S202 of the present application, the above-mentioned image to be processed may be an image (e.g., RGB image) acquired by an image acquisition device (e.g., a camera), a mobile terminal or other device. The above-mentioned multiple images to be processed may be images acquired from different viewing angles of the human body to be measured, and may include the same human body to be measured, for example, a front image and a side image of the human body to be measured. The above-mentioned human body to be measured may be a user whose size data needs to be determined, and may be a human body, for example, a consumer on an e-commerce platform. It should be noted that the above is only an example, and there is no specific limitation on the type of the image to be processed and the human body to be measured.
[0057] Optionally, this embodiment can obtain the images to be processed displayed at different viewing angles by using an image acquisition device such as a camera to obtain multiple images to be processed, and different images to be processed can be used to characterize different display states of the human body to be measured at different viewing angles. The different viewing angles may include a frontal viewing angle, a side viewing angle, a rear viewing angle, etc., which are only used as examples, and the types of viewing angles are not specifically limited.
[0058] For example, a plurality of images to be processed (e.g., RGB images) may be obtained by capturing images using a device including a camera. The plurality of images to be processed may include two perspectives, namely, the front and side perspectives of the human body to be measured, or may also include additional perspectives in addition to the front and side perspectives as multiple inputs. It should be noted that the number of input images to be processed may be selected based on actual conditions. For example, the number of input images to be processed may be selected based on the processing time. No specific limitation is imposed on the number of images to be processed herein.
[0059] Step S204, reconstructing the human body to be measured in the processed image to obtain a human body model of the human body to be measured.
[0060] In the technical solution provided in the above step S204 of the present application, the above human body model may include a human body geometric mesh, which may be used to simulate the geometric outline of the human body to be measured. The geometric outline may be the shape and contour lines of the human body to be measured in space, which may be used to characterize the three-dimensional (3D) human body shape of the human body to be measured, and may be used to determine the size of the human body to be measured, which may be a three-dimensional size.
[0061] Optionally, a plurality of images to be processed are obtained, and a human body model of the human body to be measured can be obtained by reconstructing the human body to be measured in the plurality of images to be processed, and the human body model can be used to simulate the geometric contour of the human body to be measured. It should be noted that, in this embodiment, only the human body to be measured in the object to be processed needs to be reconstructed, and other objects in the object to be processed do not need to be reconstructed.
[0062] Step S206, measuring at least one part model of the human body model to obtain size data of the part model.
[0063] In the technical solution provided in the above step S206 of the present application, the human body model may include at least one part model. The part model may be used to simulate the geometric contour of the corresponding part of the human body to be measured, and may be a component of the human body model, for example, it may be a model corresponding to the arm of the human body to be measured, a model corresponding to the upper body, a model corresponding to the lower limbs, etc. The above part model may be a component object of the human body, for example, it may be data of the abdomen, limbs, head, etc. of the human body, which is only for illustration here, and there is no specific limitation on the type of part model.
[0064] Optionally, the human body to be measured in the plurality of images to be processed is reconstructed to obtain a human body model of the human body to be measured, and the human body model may include at least one part model for simulating the geometric contour of the corresponding part of the human body to be measured. At least one part model may be detected to obtain size data of the part model. The size data may be human body data or body data, which may be used to determine the human body size information of the human body to be measured, and may be used to describe the shape, length, width and other information of the sub-object, for example, it may include information such as waist circumference, shoulder width, limb length, etc. It should be noted that this is only for example description, and the type of size data is not specifically limited.
[0065] Optionally, the part model may correspond to an actual human body part, and the feature data corresponding to the part model may be size data (i.e., the size of the part). If there are multiple part models, for example, multiple part models correspond to the shoulders, legs, and waist of an actual human body, respectively, then the corresponding size data may be shoulder width, leg length, and waist circumference, respectively.
[0066] For example, assuming that the part model is the shoulder model, leg model, and waist model in the human body model, the size data of the part model is determined as the predicted size data of the corresponding actual part of the human body. The shoulder width of the shoulder model can be determined as the predicted shoulder width of the actual human shoulder, the leg length of the leg model can be determined as the predicted leg length of the actual leg, and the waist circumference of the waist model can be determined as the predicted waist circumference of the actual waist.
[0067] Step S208, determining the size data of the part model as the predicted size data of the corresponding part.
[0068] In the technical solution provided in the above step S208 of the present application, the above predicted size data may be human size information, human size, or three-dimensional size.
[0069] In this embodiment, the image to be processed is used as input data, and the human body to be measured in the image to be processed is directly reconstructed to obtain a human body model. By measuring the human body model, the size data of the human body to be measured can be obtained, and the size data of the part model is determined as the predicted size data of the corresponding part. That is, the method quickly realizes high-precision human body size measurement based on multiple images to be processed, avoiding the high cost and low coverage caused by the traditional human body measurement relying on on-site measurement by professionals or using professional scanning equipment for measurement. At the same time, a human body model is constructed, and the part model in the human body model is directly monitored to achieve the purpose of improving the accuracy of determining the size data of the human body to be measured, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0070] For example, multiple images to be processed (e.g., full-body photos) uploaded by the human body to be measured at different viewing angles are identified. These images to be processed can be taken from different angles and can be used to show the full-body outline of the human body to be measured. Image reconstruction technology, such as 3D modeling technology, can be used to reconstruct a human body model (e.g., a 3D body model) of the human body to be measured from the uploaded multiple images to be processed. The human body model can be used to simulate the geometric outline of the human body to be measured, such as height, body shape and other information. In at least one part model of the human body model, the part model of a specific part can be further analyzed to determine the size data of the part model, such as chest circumference, waist circumference, hip circumference and other data. Optionally, this embodiment determines the size data of the part model as the predicted size data of the corresponding part, and the predicted size data can be used to further recommend the size of clothing products that match the predicted size data, such as clothing size or shoe size.
[0071] Through the above steps S202 to S208, multiple images to be processed are identified, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the human body to be measured in the images to be processed is reconstructed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; at least one part model in the human body model is measured to obtain the size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as the predicted size data of the corresponding part. That is, in the embodiment of the present application, multiple images to be processed can be obtained, for example, a front human body image and a side human body image, and a human body model of the human body to be measured (which can be a human body geometric mesh) is reconstructed using the multiple images to be processed, and the size data of the human body to be measured (for example, human body size information) is obtained by measuring the human body model, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0072] The above method of this embodiment is further introduced below.
[0073] As an optional implementation, step S206, measuring at least one part model in the human body model to obtain size data of the part model, includes: determining at least one geometric element of the part model in the human body model, wherein the geometric element is used to constitute the surface of the part model; measuring the geometric element of the part model to obtain the size data of the part model.
[0074] In this embodiment, the geometric elements may be data such as points and cross-sections of the part model, and may be used to form the surface of the part model. It should be noted that this is only an example, and the type of geometric elements is not specifically limited.
[0075] Optionally, the human body to be measured in multiple images to be processed is reconstructed to obtain a human body model of the human body to be measured, and at least one geometric element on the part model in the human body model can be determined. For example, the geometric element can be two points on the part model or a cross-section. By measuring the geometric element, the size data of the part model can be obtained.
[0076] In this embodiment, at least one geometric element of the part model in the human body model is determined to determine the size data of the part model, thereby achieving the purpose of obtaining the size of any part of the human body to be measured.
[0077] As an optional implementation, measuring geometric elements of the part model to obtain size data of the part model includes: calling a size measurement model to measure the geometric elements of the part model to obtain the size data of the part model, wherein the size measurement model is trained based on size data samples of human body samples.
[0078] In this embodiment, a size measurement model may be pre-constructed. After determining at least one geometric element of the sub-human model, the size measurement model may be called to measure the geometric element of the sub-human model to obtain size data of the sub-human model. The size measurement model may be a differentiable measuring device, which may be obtained by training size data samples based on human body samples. The size data samples may be human size data sets, for example, data sets collected by a size measurement tool, which may include but are not limited to data such as height, weight, waist circumference, and hip circumference. It should be noted that this is only an example, and there is no specific limitation on the way to obtain the size data samples.
[0079] Optionally, a large amount of open source data may be collected in advance to obtain size data samples, and a size measurement model may be trained using the size data samples. After determining at least one geometric element of the sub-human body model, the trained size measurement model may be called to measure the geometric elements of the sub-human body model to obtain the size data of the sub-human body model.
[0080] Since the human body dimensions collected manually are more reliable and accurate than the human body dimension key points marked manually, in this embodiment, the dimension data samples are used to train a dimension measurement model, and the geometric elements of the sub-human body model can be accurately measured through the dimension measurement model to accurately obtain the dimension data of the sub-human body model.
[0081] For example, a differential measuring device can be used to measure the sub-human model to determine the measurement results of geometric elements such as the straight line length, geodesic distance or perimeter of a cross section of at least one or two points of the sub-human model, thereby obtaining the size data of the sub-human model. For example, the perimeter of the cross section can be the length of the waistline of the human body to be measured, and the straight line length of the two points can be the arm length of the human body to be measured.
[0082] As an optional implementation, calling the size measurement model to measure the geometric elements of the human body model to obtain the size data of the sub-human body model includes: calling the forward propagation network in the size measurement model to measure the geometric elements of the human body model to obtain the size data of the sub-human body model.
[0083] In this embodiment, the size measurement model may include a forward propagation network. The forward propagation network in the size measurement model may be called to measure geometric elements of the human body model using the forward propagation network to obtain size data of the part model.
[0084] For example, a constructed human body model is obtained, and feature extraction can be performed on the part model in the human body model to obtain geometric elements. The extracted geometric elements are input into the forward propagation network in the trained size measurement model to perform forward propagation calculations. The forward propagation network will output predicted size data based on the input geometric elements, thereby obtaining size data of the part model, and further obtaining predicted size data of the human body to be measured. For example, the predicted size data may be data such as height, chest circumference, waist circumference, etc. of the human body to be measured.
[0085] In this embodiment, the geometric elements of the human body model are measured by calling the forward propagation network, so that accurate size data of the human body model can be obtained while improving the speed of data processing.
[0086] As an optional implementation, reconstructing the human body to be measured in the processed image to obtain a human body model of the human body to be measured includes: calling a reconstruction model to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the reconstruction model is trained based on a human body model sample of a human body sample, and the human body model sample is used to simulate the geometric contour of the human body sample.
[0087] In this embodiment, the reconstruction model may be a three-dimensional human body reconstruction module (3D Body Re-constructor), which may be a deep convolution model, and may be used to generate a human body model.
[0088] Optionally, one or more images to be processed of the same human body to be measured can be used as input data and input into the reconstruction model. The reconstruction model reconstructs the human body to be measured in the input images to be processed to obtain a human body model of the human body to be measured. The human body model can be used to characterize the geometric contour of the human body to be measured and can be represented by a mesh.
[0089] For example, a measurement application based on augmented reality technology (AR Measurement, referred to as ARM) can be used to collect human body model samples for simulating the geometric contours of human body samples, and the human body model samples based on the human body samples can be trained to obtain a reconstruction model. After obtaining the image to be processed, the reconstruction model can be called to reconstruct the image to be processed to obtain a human body model of the human body to be measured.
[0090] As an optional implementation, the method may further include: determining the degree of loss of the predicted size data compared to the actual size data of the part; determining a gradient based on the loss and the geometric element, wherein the gradient is used to represent the degree of influence of the geometric element on the loss; propagating the gradient using a back propagation network in the size measurement model; and using the propagated gradient to perform supervised training on the reconstruction model to obtain a trained reconstruction model.
[0091] In this embodiment, in the process of obtaining a reconstruction model by training a human model sample based on a human body sample, the predicted size data corresponding to the part in the human body sample can be determined, the loss degree of the predicted size data compared with the real size data of the part can be determined, and the gradient can be determined based on the loss degree and the geometric element. The gradient can be propagated by using the back propagation network in the size measurement model; the reconstruction model can be supervised and trained by using the back propagated gradient to obtain the trained reconstruction model.
[0092] Optionally, the above-mentioned real size data can be the size data obtained by actually collecting the parts. The above-mentioned gradient can be used to indicate the influence of geometric elements on the loss degree, and can be used to describe the sensitivity of the loss function to the input variable (that is, the image to be processed). The back propagation network can be used to reversely propagate from the output layer back to the input layer according to the gradient of the loss function to calculate the contribution of each neuron to the loss.
[0093] In this embodiment, a size measurement model and a reconstruction model are integrated, wherein the size measurement model may include a forward propagation network and a back propagation network. The input human body model may be processed by the forward propagation network to obtain size data of the sub-human body model. The reconstruction model may be supervised and trained by the back propagation network to obtain a trained reconstruction model. During the model training process, the model parameters of the size measurement model may remain unchanged.
[0094] In this embodiment, easily collected size data samples are used as supervisory signals for model training to reconstruct an accurate human body geometry mesh (i.e., a human body model), and then explicit measurement (e.g., explicit measurement by measurer) is used to obtain size data. End-to-end training avoids local errors in staged solutions (e.g., first measuring two-dimensional key points, then regressing key points to size). Moreover, as supervisory information, manually collected size data samples are more reliable and accurate than manually labeled human body size key points. Furthermore, a human body model is reconstructed through a trained reconstruction model, and the trained size measurement model is used to measure the part model in the human body model to obtain size data, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0095] For example, during the model training process, the predicted size data obtained by measuring the human body model output by the reconstruction model using the size measurement model can be compared, the loss degree of the predicted size data and the real size data can be compared, and the gradient can be determined based on the loss degree and the geometric elements. The gradient can be propagated using the back propagation network in the size measurement model; the back propagated gradient can be used to complete the supervised training process of the reconstruction model to obtain the trained reconstruction model, thereby achieving the purpose of improving the accuracy of the human body model generated by the reconstruction model. In the mode use stage, the image to be processed can be used as input, the trained reconstruction model can be used to process the image to be processed, the human body model can be reconstructed, and the size measurement model can be used to perform geometric measurement on the human body model to obtain the size data of the part model.
[0096] As an optional implementation, a reconstruction model is called to reconstruct a human body to be measured in an image to be processed to obtain a human body model of the human body to be measured, including: using a feature extraction module in the reconstruction model to extract initial image features of the human body to be measured from the image to be processed; using a feature connection module in the reconstruction model to connect multiple initial image features corresponding to multiple images to be processed to obtain target image features; using a regression processing module in the reconstruction model to perform regression processing on the target image features to obtain geometric parameters of the human body to be measured, and generating a human body model based on the geometric parameters.
[0097] In this embodiment, the reconstruction model may include a feature extraction module (enconder), a feature connection module (concat) and a regression processing module (regressor). Among them, the above-mentioned feature extraction module can be an encoder module, which can be used to extract the initial image features of the human body to be measured from the image to be processed, and can be used to convert the image to be processed into a vector representation. The above-mentioned feature connection module can be used to splice multiple features, which can be used to increase the representation ability of the reconstruction model. The above-mentioned regression processing module can be a regressor, which can be used for regression tasks to generate a human body model.
[0098] Optionally, an image to be processed is obtained, and the image to be processed is transmitted to a reconstruction model as input data. The reconstruction model can be called to reconstruct the image to be processed through the following steps to obtain a human body model of the human body to be measured: the feature extraction module in the reconstruction model can be used to extract the initial image features of the human body to be measured from the image to be processed (for example, features (feat1, feat2)); the feature connection module in the reconstruction model is used to connect multiple initial image features corresponding to multiple images to be processed to obtain target image features (feat_both); the regression processing module in the reconstruction model is used to perform regression processing on the target image features to obtain geometric parameters of the human body to be measured, and a human body model is generated based on the geometric parameters. The geometric parameters can be human body parameters, for example, they can be body shape coefficients of a skinned multi-person linear model (SMPL for short).
[0099] For example, two images to be processed are obtained, and the images to be processed are transmitted as input data to the reconstruction model. The encoder in the reconstruction model extracts features from the two images to be processed respectively to obtain initial image features (e.g., feat1, feat2), and the initial image features are cascaded into target image features (feat_both) along the channel direction in the feature connection module. The target image features are then input into the regressor to regress a set of 100-dimensional SMPL body shape coefficients, and finally obtain a human body model (e.g., human body geometric grid). It should be noted that the size of the above numbers is only for example and is not specifically limited here.
[0100] Optionally, different dimensions of the SMPL body shape coefficient can correspond to different dimensional information. For example, the first dimension can be used to control height, and the larger the value, the taller the height; the second dimension can be used to control fatness and thinness, and the larger the value, the thinner the person; the fifth dimension can be used to control the size of the hips, and the eighth dimension can be used to control the thickness of the calves, etc. The more dimensions, the richer the body shape changes.
[0101] As an optional implementation, generating a human body model based on geometric parameters includes: determining multiple vertices of a geometric contour of a human body to be measured based on the geometric parameters; and creating a three-dimensional geometric mesh from the vertices, wherein the human body model includes the three-dimensional geometric mesh.
[0102] In this embodiment, after obtaining the geometric parameters of the human body to be measured, multiple vertices of the geometric contour of the human body to be measured can be determined based on the geometric parameters, and the multiple vertices can be created as a three-dimensional geometric mesh, for example, multiple vertices can be connected to construct a three-dimensional set mesh. The three-dimensional geometric mesh can be a human body geometric mesh.
[0103] As an optional implementation, based on geometric parameters, multiple vertices of the geometric contour of the human body to be measured are determined, including: deviating the vertices of the reference geometric contour according to the geometric parameters to obtain the vertices of the geometric contour of the human body to be measured.
[0104] In this embodiment, the geometric parameters of the human body to be measured are obtained, and according to the geometric parameters, the vertices of the reference geometric contour can be deviated to obtain the vertices of the geometric contour of the human body to be measured. The reference geometric contour can be a pre-set geometric contour, a geometric contour corresponding to a human body template obtained by SMPL through statistics, or a geometric contour corresponding to an average body shape.
[0105] Optionally, the vertices of the geometrical contour of the human body to be measured can be obtained by the following steps:
[0106] The vertex of the geometric contour = templ ate + bas is * coefficients
[0107] Wherein, template can be a human body template obtained by SMPL through statistics, which can represent the average body shape. Bases can be a body shape basis obtained by SMPL through statistics. Coefficients can be the predicted body shape coefficients (i.e., geometric parameters) of the human body to be measured in the image to be processed. Bases*coefficients represents the change in the body shape of the human body to be measured relative to the average body shape.
[0108] As an optional implementation, the human body samples include human body samples wearing clothing products of different sizes.
[0109] In this embodiment, in order to increase the generalization ability of the reconstruction model and the size measurement model, human body samples wearing clothing products of different sizes can be obtained, and the size measurement model can be trained using the size data samples corresponding to the human body samples, and the reconstruction model can be trained using the human body model samples corresponding to the human body samples.
[0110] Optionally, when constructing human body samples, multi-view images of each model wearing clothing products of different tightness, such as tight, fitted, and loose, can be collected to further increase the generalization ability of the reconstruction model and the size measurement model to process clothing products of different sizes.
[0111] As an optional implementation, identifying multiple images to be processed includes: identifying a front image to be processed and a side image to be processed, wherein the human body to be measured is displayed in the front image to be processed from a front perspective, and the human body to be measured is displayed in the side image to be processed from a side perspective.
[0112] In this embodiment, the front image to be processed and the side image to be processed in the image to be processed can be obtained, and the predicted size data of the human body to be measured can be accurately determined by processing the front image to be processed and the side image to be processed. The side view angle can include the left side or the right side, and the side view angle is not specifically limited here.
[0113] In the embodiment of the present application, a plurality of images to be processed may be obtained, for example, a front human body image and a side human body image, and a human body model (which may be a human body geometry mesh) of the human body to be measured may be reconstructed using the plurality of images to be processed. By measuring the human body model, the size data of the human body to be measured (for example, human body size information) may be obtained, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0114] The embodiment of the present application also provides a method for measuring the size data of a biological object from the perspective of measuring the size of a human body. Figure 3 is a flow chart of a method for measuring size data of a biological object according to an embodiment of the present application, such as Figure 3 As shown, the method may include the following steps.
[0115] Step S302: identifying a plurality of images to be processed.
[0116] In the technical solution provided in the above step S302 of the present application, the above biological objects may include humans, animals and the like, which is only used as an example and does not impose any specific limitation on the type of biological objects.
[0117] For example, multiple images may be acquired, and the biological objects contained in different images may be the same or different. Graphics including the same biological objects may be determined as images to be processed, so as to identify and obtain multiple images to be processed.
[0118] Step S304: reconstruct the biological object to be measured in the processed image to obtain a biological object model of the biological object to be measured.
[0119] In the technical solution provided in the above step S304 of the present application, the above biological human body model may include a geometric grid of the biological object, which may be used to simulate the geometric outline of the biological object to be measured. The geometric outline may be used to characterize the shape and contour lines of the biological object to be measured in a multidimensional space, may be used to characterize the three-dimensional body shape of the biological object to be measured, and may be used to determine the three-dimensional size of the biological object to be measured.
[0120] Optionally, a plurality of images to be processed are acquired, and a biological object model of the biological object to be measured can be obtained by reconstructing the plurality of images to be processed. The biological object model can be used to simulate the geometrical contour of the biological object to be measured.
[0121] Step S306, measuring at least one part model in the biological object model to obtain size data of the part model.
[0122] In the technical solution provided in the above step S306 of the present application, the biological object model includes at least one part model. For example, when the biological object model is a cat, the part model may be a model corresponding to the cat's limbs, body, etc. The above part model may be used to simulate the geometric contour of the corresponding part in the biological object to be measured.
[0123] Optionally, after reconstructing a plurality of images to be processed, a biological object model of the biological object to be measured is obtained, and the biological object model may include at least one part model for simulating the geometric contour of the corresponding part of the biological object to be measured. The at least one part model may be measured to obtain size data of the part model. The size data may be the body data of the organism, which may be used to determine the body size data of the biological object to be measured, for example, it may be used to describe the shape, length, width and other information of the part. It should be noted that this is only for example description, and the type of size data is not specifically limited.
[0124] Step S308, determining the size data of the part model as the predicted size data of the corresponding part.
[0125] In the technical solution provided in the above step S308 of the present application, the size data of the part model can be determined as the predicted size data of the part in the biological object to be measured. For example, when the part model is used to represent the geometric contour of the arm of the biological object to be measured, the size data of the part model can be determined as the predicted size data corresponding to the arm.
[0126] In an embodiment of the present application, a plurality of images to be processed are identified, wherein the image contents of the plurality of images to be processed include the same biological object to be measured, and the biological object to be measured is displayed in different images to be processed at different viewing angles; the biological object to be measured in the images to be processed is reconstructed to obtain a biological object model of the biological object to be measured, wherein the biological object model is used to simulate the geometric contour of the biological object to be measured; at least one part model in the biological human body model is measured to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the biological object to be measured; the size data of the part model is determined as the predicted size data of the corresponding part, thereby achieving the technical effect of improving the efficiency of measuring the size data of the object and solving the technical problem of low efficiency in measuring the size data of the object.
[0127] The embodiment of the present application is directed to a usage scenario, such as a size recommendation scenario, and further provides a method for recommending product information, which can be applied to an e-commerce platform. The e-commerce platform can be used to provide a convenient size measurement tool, which can be used to determine product attribute information of a product to be recommended based on the determined predicted size data. Figure 4 FIG. 1 is a flow chart of a method for recommending product information according to an embodiment of the present application. Figure 4 As shown, the method may include the following steps:
[0128] Step S402: identifying a plurality of images to be processed.
[0129] In the technical solution provided in the above step S402 of the present application, the image contents of the multiple images to be processed may include the same human body to be measured, and the human body to be measured may be displayed in different images to be processed from different perspectives.
[0130] Step S404: reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured.
[0131] In the technical solution provided in the above step S404 of the present application, the above human body model can be used to simulate the geometric contour of the human body to be measured.
[0132] Step S406, measuring at least one part model of the human body model to obtain size data of the part model.
[0133] In the technical solution provided in the above step S406 of the present application, the above human body model may include at least one part model, and the above part model may be used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0134] Step S408, determining the size data of the part model as the predicted size data of the corresponding part.
[0135] Step S410, recommending product attribute information that matches the predicted size data, wherein the clothing products that meet the product attribute information are to be worn by the corresponding parts, and the clothing products come from a virtual clothing store deployed on the e-commerce platform.
[0136] In the technical solution provided in the above step S410 of the present application, after determining the predicted size data of the part, product attribute information matching the predicted size data can be recommended to the user (e.g., the purchaser) based on the predicted size data. The above product attribute information can be the size information, length information, etc. of the product, for example, it can be information such as large size, small size, medium size, etc., or it can be information such as 35 cm, 25 cm, etc. The above products can be information such as clothing products, household items, etc. The above clothing products can include clothing, accessories, etc., which can be commodities selected by the purchaser. This is only for example, and there is no specific restriction on the type and determination method of the clothing products.
[0137] Optionally, there may be a virtual clothing store in the e-commerce platform. When a user purchases clothing products in the virtual clothing store, multiple images to be processed may be uploaded to the e-commerce platform. The e-commerce platform obtains the images to be processed and reconstructs the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured in the images to be processed. At least one part model in the human body model may be measured to obtain the size data of the part model. The size data of the part model may be determined as the predicted size data of the part. Based on the predicted size data, the product attribute information of the clothing products suitable for the part in the virtual clothing store may be recommended to the user and recommended to the user. The clothing products with the final determined product attribute information can be appropriately worn on the part.
[0138] Optionally, different types of products can be sold in an e-commerce platform, for example, daily products such as tablecloths and bed sheets, or clothing products such as clothes, shoes, and pants can be sold. However, it is impossible to accurately determine the attribute information of the product to be purchased, such as the size and length of the product. To solve the above problem, in this embodiment, a human body model is reconstructed, and the size data of the part model is obtained by measuring the human body model, and the predicted size data of the part can be further determined. Based on the predicted size data, the attribute information of the product to be recommended to the user can be accurately determined.
[0139] In an embodiment of the present application, a plurality of images to be processed are identified, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the human body to be measured in the images to be processed is reconstructed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; at least one part model in the human body model is measured to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as predicted size data of the corresponding part; and product attribute information matching the predicted size data is recommended, wherein the clothing product satisfying the product attribute information is to be worn by the corresponding part, and the clothing product comes from a virtual clothing store deployed on an e-commerce platform, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0140] The embodiment of the present application also provides another method for recommending product information, which can be applied to a client, and the client is used to deploy a virtual clothing store on an e-commerce platform. Figure 5 FIG. 1 is a flow chart of another method for recommending product information according to an embodiment of the present application. Figure 5 As shown, the method may include the following steps:
[0141] Step S502: identifying a plurality of images to be processed.
[0142] In the technical solution provided in the above step S502 of the present application, the image content of the above image to be processed may include the same human body to be measured, and the human body to be measured may be displayed in different images to be processed from different perspectives.
[0143] Step S504: reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured.
[0144] In the technical solution provided in the above step S504 of the present application, the above human body model is used to simulate the geometric contour of the human body to be measured.
[0145] Step S506, measuring at least one part model of the human body model to obtain size data of the part model.
[0146] In the technical solution provided in the above step S506 of the present application, the above part model can be used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0147] Step S508, determining the size data of the part model as the predicted size data of the corresponding part.
[0148] Step S510: recommending product attribute information matching the predicted size data to the e-commerce platform.
[0149] In the technical solution provided in the above step S510 of the present application, the client can be a merchant of a virtual clothing store in an e-commerce platform. The merchant can obtain the image to be processed, reconstruct the image to be processed to obtain a human body model of the human body to be measured in the image to be processed, measure at least one part model in the human body model to obtain the size data of the part model, and determine the size data of the part model as the predicted size data of the part. Based on the predicted size data, the product attribute information of the clothing products suitable for the part in the virtual clothing store can be determined and recommended to the e-commerce platform, so that the clothing products suitable for the buyer can be recommended to the user through the e-commerce platform. Among them, the clothing products that meet the product attribute information can be worn by the corresponding part, and the clothing products can come from the virtual clothing store.
[0150] In an embodiment of the present application, a plurality of images to be processed are identified, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the human body to be measured in the images to be processed is reconstructed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; at least one part model in the human body model is measured to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as predicted size data of the corresponding part; product attribute information matching the predicted size data is recommended to an e-commerce platform, wherein a clothing product that meets the product attribute information is to be worn by the corresponding part, and the clothing product comes from a virtual clothing store, thereby achieving a technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0151] The embodiment of the present application also provides another method for measuring human body size from the perspective of human-computer interaction. Figure 6 FIG. 1 is another flow chart of measuring the size of a human body according to an embodiment of the present application. Figure 6 As shown, the method may include the following steps:
[0152] Step S602 , in response to an image input operation on the operation interface, a plurality of images to be processed are displayed on the operation interface.
[0153] In the technical solution provided in the above step S602 of the present application, the above image input operation can be triggered by the user, and is used to input multiple images to be processed on the operation interface, so that the embodiment responds to the image input operation on the operation interface and displays multiple images to be processed on the operation interface. The image content of the above multiple images to be processed may include the same human body to be measured, and the human body to be measured can be displayed in different images to be processed from different perspectives.
[0154] Step S604 , in response to the reconstruction operation performed on the operation interface, a human body model of the human body to be measured is displayed on the operation interface.
[0155] In the technical solution provided in the above step S604 of the present application, the reconstruction operation can be triggered by the user to reconstruct the image to be processed to obtain a human body model, and display the human body model of the human body to be measured on the operation interface. The human body model can be reconstructed from the human body to be measured in the image to be processed, and is used to simulate the geometric contour of the human body to be measured.
[0156] Step S606, in response to the size measurement operation performed on the operation interface, the predicted size data of at least one part of the human body to be measured is displayed on the operation interface.
[0157] In the technical solution provided in the above step S606 of the present application, the size measurement operation can be triggered by the user to measure the part model in the human body model to determine the size data of at least one part model, and the size data of at least one part model can be determined as the predicted size data of the part, and the finally determined predicted size data can be displayed in the operation interface. Among them, the predicted size data of the above part is determined based on the size data of the corresponding part model in the human body model, and the part model is used to simulate the geometric contour of the corresponding sub-object.
[0158] In an embodiment of the present application, in response to an image input operation on an operation interface, a plurality of images to be processed are displayed on the operation interface, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; in response to a reconstruction operation on the operation interface, a human body model of the human body to be measured is displayed on the operation interface, wherein the human body model is obtained by reconstructing the human body to be measured in the image to be processed and is used to simulate the geometric contour of the human body to be measured; in response to a size measurement operation on the operation interface, predicted size data of at least one part of the human body to be measured is displayed on the operation interface, wherein the predicted size data of the part is determined based on the size data of a corresponding part model in the human body model, and the part model is used to simulate the geometric contour of the corresponding part, thereby achieving a technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0159] According to an embodiment of the present application, a method for measuring the size data of an object in a virtual reality scene such as a virtual reality device and an augmented reality device is also provided. Figure 7 is a flow chart of measuring the size data of another object according to an embodiment of the present application. Figure 7 As shown, the method may include the following steps:
[0160] Step S702: Display multiple images to be processed on a presentation screen of a virtual reality (VR) device or an augmented reality (AR) device, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives.
[0161] In the technical solution provided in the above step S702 of the present application, the image input instruction on the operating interface of the virtual reality device or the augmented reality device can be triggered by the image acquisition device or the user. In response to the image input instruction, multiple images to be processed can be displayed on the presentation screen of the VR device or the AR device.
[0162] Step S704: drive the VR device or the AR device to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured.
[0163] Optionally, the inquiry information to be answered is obtained, and the dialogue model can be called to answer the inquiry information to obtain initial answer information.
[0164] Step S706: driving the VR device or the AR device to measure at least one part model of the human body model to obtain size data of the part model.
[0165] In the technical solution provided in the above step S706 of the present application, multiple images to be processed can be displayed on the presentation screen of the VR device or AR device, the VR device or AR device can be driven to reconstruct the model of the images to be processed to obtain a human body model of the human body to be measured, and the VR device or AR device can be driven to measure at least one part model in the human body model to obtain the size data of the part model. The above part model is used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0166] In step S708, the size data of the part model is determined as the predicted size data of the corresponding part, and the predicted size data of the part is displayed on the presentation screen of the VR device or the AR device.
[0167] Optionally, in this embodiment, the size data measurement method of the object can be applied to a hardware environment composed of a server and a virtual reality device. The predicted size data of the part is displayed on the presentation screen of the VR device or AR device. The server can be a server corresponding to an e-commerce platform or a client. The network includes but is not limited to: a wide area network, a metropolitan area network or a local area network. The virtual reality device is not limited to: a virtual reality helmet, virtual reality glasses, a virtual reality all-in-one machine, etc.
[0168] Optionally, the virtual reality device includes: a memory, a processor, and a transmission device. The memory is used to store an application program, which can be used to execute: displaying multiple images to be processed on a presentation screen of a virtual reality VR device or an augmented reality AR device, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; driving the VR device or AR device to reconstruct the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; driving the VR device or AR device to measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determining the size data of the part model as the predicted size data of the corresponding part, and displaying the predicted size data of the part on the presentation screen of the VR device or AR device.
[0169] It should be noted that the object processing method applied in the VR device or AR device in the embodiment may include: Figure 3 The method of the illustrated embodiment is used to achieve the purpose of driving the predicted size data of the display part of the VR device or AR device.
[0170] Optionally, the processor of this embodiment can call the application stored in the memory to execute the above steps through a transmission device. The transmission device can receive media files sent by the server through the network, and can also be used for data transmission between the processor and the memory.
[0171] Optionally, in a virtual reality device, a head-mounted display with eye tracking is provided, wherein the screen in the HMD is used to display the displayed video images, the eye tracking module in the HMD is used to obtain the real-time movement trajectory of the user's eyes, the tracking system is used to track the user's position information and motion information in the real three-dimensional space, and the computing processing unit is used to obtain the user's real-time position and motion information from the tracking system, and calculate the three-dimensional coordinates of the user's head in the virtual three-dimensional space, as well as the user's field of view direction in the virtual three-dimensional space, etc.
[0172] In an embodiment of the present application, a virtual reality device may be connected to a terminal, and the terminal and the server are connected through a network. The virtual reality device is not limited to: a virtual reality helmet, virtual reality glasses, a virtual reality all-in-one machine, etc. The terminal is not limited to a PC, a mobile phone, a tablet computer, etc. The server may be a server corresponding to a media file operator, and the network includes but is not limited to: a wide area network, a metropolitan area network or a local area network.
[0173] Figure 8 is a schematic diagram of predicted size data of a sub-object according to an embodiment of the present application, such as Figure 8 As shown, multiple images to be processed, such as image 801 to be processed and image 802 to be processed, can be displayed on a presentation screen of a virtual reality VR device or an augmented reality AR device, and the VR device or the AR device can be driven to reconstruct the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; the VR device or the AR device can be driven to measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as the predicted size data of the corresponding part, and the predicted size data of the part is displayed on the presentation screen of the VR device or the AR device, such as shoulder width: 48 cm, chest circumference: 48 cm, hip circumference: 30 cm, and waist circumference: 97 cm.
[0174] In an embodiment of the present application, a plurality of images to be processed are displayed on a presentation screen of a virtual reality VR device or an augmented reality AR device, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the VR device or the AR device is driven to reconstruct the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; the VR device or the AR device is driven to measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as the predicted size data of the corresponding part, and the predicted size data of the part is displayed on the presentation screen of the VR device or the AR device, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0175] According to an embodiment of the present application, an embodiment of an image processing system is also provided. It should be noted that the image processing system of this embodiment can be used to execute the object size data measurement method of the embodiment of the present invention. Fig. 9 is a schematic diagram of an image processing system according to an embodiment of the present application, such as Fig. 9 As shown, the image processing system 900 may include: a client 902 and a server 904 .
[0176] Client 902 uploads multiple images to be processed.
[0177] In this embodiment, multiple images to be processed may be uploaded via the client 902. The image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles.
[0178] The server 904 is used to reconstruct the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determine the size data of the part model as predicted size data of the corresponding part; and send the predicted size data of the part to the client.
[0179] In this embodiment, the server 904 can obtain multiple images to be processed uploaded by the client 902, and reconstruct the images to be processed to obtain a human body model of the human body to be measured. Furthermore, at least one part model in the human body model can be measured to obtain size data of the part model; the size data of the part model can be determined as predicted size data of the corresponding part; and the predicted size data of the part can be sent to the client.
[0180] In this embodiment, a plurality of images to be processed are uploaded through the client 902, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; the server 904 is used to reconstruct the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; at least one part model in the human body model is measured to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; the size data of the part model is determined as the predicted size data of the corresponding part; and the predicted size data of the part is sent to the client, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0181] Currently, consumers have a weak sense of their own size data, and manual measurement and input costs are high. In addition, some clothing orders are refunded due to size mismatch, which brings huge refund costs.
[0182] In the related art, the three-dimensional size of the human body to be measured is usually automatically measured based on images taken by ordinary mobile phones (for example, RGB images), and then a size file is created. For consumers, this method can solve the pain points of not knowing their own size and difficulty in choosing clothing sizes when shopping online. For e-commerce platforms, a large amount of consumer size data can be quickly accumulated at a low cost, and size recommendations and size shopping guides can be made based on these data, thereby improving transactions and conversions, and reducing refunds and returns due to size reasons.
[0183] As an optional example, high-end instruments, such as 3D scanners, can be used to improve the measurement accuracy of human body size. However, this method is also costly and is not suitable for promotion to users on the consumer side (C-side). Alternatively, RGB images can be used as input visual solutions, such as pure 2D solutions (human key point measurement or 2D + 3D hybrid solutions (such as 2D key point measurement + 3D human body shape feature extraction). However, the above methods still have the technical problem of low measurement accuracy.
[0184] As another optional example, images taken by the user from the front and side are obtained, and the images are processed to obtain body data of multiple parts of the body, such as waist circumference, shoulder width, limb length, etc., through measurement and calculation. Further, the size suitable for the user can be recommended based on the calculated body data, and the suitable color and style can be recommended based on the user's appearance, age and other information. However, the measurement accuracy of this method is seriously affected by the looseness of the user's clothing, and the measurement error is high and the variance is large, and there is a technical problem of low efficiency in measuring the size data of the human body.
[0185] In order to process the images to be measured taken by consumer-grade devices (such as mobile phone cameras) to quickly achieve high-precision human body size measurement, and avoid the high cost and low coverage caused by the traditional human body measurement relying on professional on-site measurement or using professional scanning equipment for measurement, in this embodiment, a human body size measurement method based on a differentiable measuring device is proposed. The method can be based on multiple images to be processed, such as front images to be processed and side images to be processed, without the need for additional depth information (such as a mobile phone with Lidar) or a high-precision human 3D model (such as a professional human body scanning device), and can achieve accurate measurement of the predicted size data of the sub-object of the human body to be measured. By using a differentiable measuring device, it is convenient to supervise the accurate reconstruction of the three-dimensional body shape of the human body with the collected human body size data, and directly measure the reconstructed human body model to obtain the size value of any part, and it is convenient to collect training data, thereby achieving the technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0186] The following is a further introduction to a method for measuring human body size based on a differentiable measuring device proposed in an embodiment of the present application.
[0187] Fig.10 is a schematic diagram of a human body size measurement framework based on a differential measuring device according to an embodiment of the present application, such as Fig.10 As shown, the framework may be composed of a 3D human body reconstruction module 1002 and a differentiable measurement module 1004 .
[0188] like Fig.10 As shown, a plurality of images to be processed 1001 may be obtained, and the plurality of images to be processed 1001 may be reconstructed by a 3D human body reconstruction module 1002 to obtain a human body model 1003. A differential measurement module 1004 may obtain the human body model 1003, and measure the human body model 1003 to determine at least one geometric element of the human body model 1003 (e.g., Fig.10 , b, c, d, e, f, p, k, m, etc.) to obtain size data 1005. Further, the predicted size data 1006 corresponding to the parts can be obtained, such as shoulder width, arm length, chest thickness, arm width, etc. Among them, the 3D human body reconstruction module 1002 can be used to reconstruct a human body model, which can be a reconstruction model. The differential measurement module 1004 can be used to measure the human body model, which can be a size measurement model.
[0189] In this embodiment, the reconstruction model may include a feature extraction module, a feature connection module and a regression processing module.
[0190] For example, two images to be processed are obtained, and the images to be processed are transmitted as input data to the reconstruction model. The encoder in the reconstruction model extracts features from the two images respectively to obtain initial image features (e.g., feat1, feat2), and the initial image features are cascaded into target image features (feat_both) along the channel direction in the feature connection module. The target image features are then input into the regressor to regress a set of 100-dimensional SMPL body shape coefficients, and finally obtain a human body model (e.g., human body geometric grid). It should be noted that the size of the above numbers is only for example and is not specifically limited here.
[0191] Optionally, a plurality of images to be processed (e.g., RGB images) may be obtained by photographing a device including a camera. The plurality of images to be processed may include two perspectives, namely, the front and side perspectives of the human body to be measured, or may also include additional perspectives in addition to the front and side perspectives as multiple inputs. It should be noted that the number of input images to be processed may be selected according to actual conditions. For example, the number of input images to be processed may be selected according to the processing time. No specific limitation is imposed on the number of images to be processed herein.
[0192] Optionally, different dimensions of the SMPL body shape coefficient can correspond to different dimensional information. For example, the first dimension can be used to control height, and the larger the value, the taller the height; the second dimension can be used to control fatness and thinness, and the larger the value, the thinner the person; the fifth dimension can be used to control the size of the hips, and the eighth dimension can be used to control the thickness of the calves, etc. The more dimensions, the richer the body shape changes.
[0193] Optionally, the vertices of the geometrical contour of the human body to be measured can be obtained by the following steps:
[0194] The vertex of the geometric contour = templ ate + bas is * coefficients
[0195] Wherein, template can be a human body template obtained by SMPL through statistics, which can represent the average body shape. Bases can be a body shape basis obtained by SMPL through statistics. Coefficients can be the predicted body shape coefficients (i.e., geometric parameters) of the human body to be measured in the image to be processed. Bases*coefficients represents the change in the body shape of the human body to be measured relative to the average body shape.
[0196] In this embodiment, the differential measurement module can use a deep learning framework (eg, a pytorch module) customized using an underlying programming language (cuda operator).
[0197] Optionally, in this embodiment, the measurement logic can be implemented according to the underlying programming language. For example, the measurement operator mainly measures two lengths along the surface of the human body model. The geodesic distance between two points (for example, arm length) can be determined as the geodesic distance measured from the shoulder point to the wrist point; the circumference of the cross-section (for example, waistline) is the circumference of the cross-section measured along the waist cross-section.
[0198] In this embodiment, the differential measurement module may include a forward propagation network and a backward propagation network, which are used for forward reasoning and backward gradient propagation, respectively. The forward propagation network can be used to measure the straight-line distance between two points on the mesh or the perimeter of a cross section to obtain size data, which can be in centimeters. It should be noted that this is only an example, and the unit of the size data is not specifically limited.
[0199] Optionally, during the training process of the differential measurement module, there are no model parameters that need to be learned.
[0200] Optionally, this embodiment can implement the gradient backpropagation of the backpropagation network in the differential measurement module by calling backward of pytorch.
[0201] In this embodiment, a large amount of open source data can be collected to train the 3D human body reconstruction module.
[0202] Optionally, Fig.11 is a schematic diagram of a data sample according to an embodiment of the present application, such as Fig.11 As shown in the figure, a large number of human size data sets can be collected. When constructing human samples, multi-view images of each model wearing clothing products of different tightness, such as tight, fitted, and loose, as well as size data samples of the model object can be collected to further increase the generalization ability of the reconstruction model and size measurement model to process clothing products of different sizes.
[0203] Optionally, during the model training process, the predicted size data obtained by measuring the part model output by the reconstruction model using the size measurement model can be compared, the loss degree of the predicted size data and the real size data can be compared, and the gradient can be determined based on the loss degree and the geometric elements. The gradient can be propagated using the back propagation network in the size measurement model; the back propagated gradient can be used to complete the supervised training process of the reconstruction model to obtain the trained reconstruction model, thereby achieving the purpose of improving the accuracy of the human body model generated by the reconstruction model. In the mode use stage, the image to be processed can be used as input, the trained reconstruction model can be used to process the image to be processed, the human body model can be reconstructed, and the size measurement model can be used to perform geometric measurement on the human body model to obtain the size data of the sub-human body model.
[0204] Optionally, the embodiment can also obtain sample data by using a human body scanning device. The scanning device can be used to perform 3D reconstruction of the human body, and then direct size measurement can be performed based on the reconstructed 3D human body model. This method has high accuracy and high cost. Alternatively, specially designed clothes can be worn to calculate the human body size based on the distribution of landmark points on the clothes.
[0205] In this embodiment, the reconstruction model and the size measurement model are integrated, and the easily collected size data samples are used as the supervisory signal for model training to reconstruct an accurate human body geometry mesh (i.e., human body model), and then explicit measurement (e.g., explicit measurement by measurer) is used to obtain size data. End-to-end training avoids local errors in staged solutions (e.g., first measuring two-dimensional key points and then regressing the key points to size). Moreover, as supervisory information, the manually collected size data samples are more reliable and accurate than the manually labeled human body size key points, thus achieving the purpose of improving the measurement accuracy of the model.
[0206] In this embodiment, the human body geometry mesh is directly reconstructed using the image as input, and then the human body size data is obtained by measuring the mesh. On the other hand, when constructing the human body sample, multi-view images of each model wearing clothing products of different tightness, such as tight, fitted, and loose, can be collected to further increase the generalization ability of the reconstruction model and the size measurement model to process clothing products of different sizes.
[0207] In this embodiment, a small AR body measurement model and a large AR body measurement model may be included, wherein the small model can be measured in real time on a mobile phone, and the large model can be deployed on the server to measure the size data of the part.
[0208] Optionally, compared with intelligent body measurement (eg, AI body measurement), the photo taking completion rate and measurement adoption rate of this embodiment are higher than those of intelligent body measurement.
[0209] In this embodiment, a differentiable measurement module is extracted, and human body geometry reconstruction is supervised using more easily collected human body size data; the entire framework is an end-to-end method with images as input, which avoids error-prone key point measurement or size regression modules and reduces the impact of clothing looseness during the measurement process.
[0210] The average measurement error of the large human body measurement model in this embodiment is 2.2 cm, which is much better than other products in terms of accuracy; the average measurement error of the small model is 2.6 cm, which is a measurement model that can be fully deployed on the terminal side.
[0211] In this embodiment, a plurality of images to be processed may be obtained, for example, a front human body image and a side human body image, and a human body model (which may be a human body geometric mesh) of the human body to be measured may be reconstructed using the plurality of images to be processed. By measuring the human body model, the size data of the human body to be measured (for example, human body size data) may be obtained, thereby achieving a technical effect of improving the efficiency of measuring the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0212] The method embodiment provided in Embodiment 1 of the present application may also be executed in a mobile terminal, a computer terminal or a similar computing device. Fig.12 is a hardware structure block diagram of a computer terminal (or mobile device) for a method for measuring size data of an object according to an embodiment of the present application, such as Fig.12 As shown, the computer terminal 120 (or mobile device) may include one or more (shown in the figure as 1202a, 1202b, ..., 1202n) processors 1202 (the processor 1202 may include but is not limited to a processing device such as a microprocessor (Microcontroller Unit, referred to as MCU) or a programmable logic device (Field Programmable Gate Array, referred to as FPGA)), a memory 1204 for storing data, and a transmission device 1206 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (Universal Ser ia l Bus, referred to as USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Fig.12 The structure shown is only for illustration and does not limit the structure of the above electronic device. Fig.12 More or fewer components as shown, or with Fig.12 Different configurations are shown.
[0213] Fig.12 The hardware structure block diagram shown can be used not only as an exemplary block diagram of the above-mentioned computer terminal 120 (or mobile device), but also as an exemplary block diagram of the above-mentioned server.
[0214] The memory 1204 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the data processing method in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 1204, that is, realizing the above-mentioned data processing method. The memory 1204 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 1204 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer terminal 120 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0215] The transmission device 1206 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by the communication provider of the computer terminal 120. In one example, the transmission device 1206 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 1206 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.
[0216] The display may be, for example, a touch screen liquid crystal display (LCD), which may enable a user to interact with a user interface of the computer terminal 120 (or mobile device).
[0217] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.
[0218] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0219] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.
[0220] According to an embodiment of the present application, there is also provided a method for implementing the above Figure 2 The object size data measuring method shown is an object size data measuring device.
[0221] Fig.13 is a schematic diagram of a human body size measuring device according to an embodiment of the present application, such as Fig.13 As shown, the human body size measurement device 1300 may include: a first recognition unit 1302 , a first reconstruction unit 1304 , a first measurement unit 1306 and a first determination unit 1308 .
[0222] The first recognition unit 1302 is configured to recognize a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles.
[0223] The first reconstruction unit 1304 is used to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured.
[0224] The first measuring unit 1306 is used to measure at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0225] The first determining unit 1308 is used to determine the size data of the part model as the predicted size data of the corresponding part.
[0226] It should be noted that the first identification unit 1302, the first reconstruction unit 1304, the first measurement unit 1306 and the first determination unit 1308 correspond to steps S202 to S208 in Embodiment 1, and the four units and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the above units may be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above units may also be part of the device and run in the computer terminal 120 provided in Embodiment 3.
[0227] According to an embodiment of the present application, another method for implementing the above Figure 3 The biological object size data measuring method and the biological object size data measuring device are shown.
[0228] Fig.14 is a schematic diagram of a biological object size data measuring device according to an embodiment of the present application, such as Fig.14 As shown, the biological object size data measuring device 1400 may include: a second recognition unit 1402 , a second reconstruction unit 1404 , a second measurement unit 1406 and a second determination unit 1408 .
[0229] The second identification unit 1402 is used to identify a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same biological object to be measured, and the biological object to be measured is displayed in different images to be processed at different viewing angles.
[0230] The second reconstruction unit 1404 is used to reconstruct the image to be processed to obtain a biological object model of the biological object to be measured, wherein the biological object model is used to simulate the geometric contour of the biological object to be measured.
[0231] The second measuring unit 1406 is used to measure at least one part model in the biological object model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the biological object to be measured.
[0232] The second determining unit 1408 is used to determine the size data of the part model as the predicted size data of the corresponding part.
[0233] It should be noted that the second identification unit 1402, the second reconstruction unit 1404, the second measurement unit 1406 and the second determination unit 1408 correspond to steps S302 to S308 in Embodiment 1, and the four units and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the above units may be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above units may also be part of the device and may be run in the computer terminal 120 provided in Embodiment 3.
[0234] According to an embodiment of the present application, another method for implementing the above Figure 4 The product information recommendation method and the product information recommendation device shown can be applied to an e-commerce platform.
[0235] Fig.15 is a schematic diagram of a device for recommending product information according to an embodiment of the present application, such as Fig.15 As shown, the device 1500 for recommending product information may include: a third identification unit 1502 , a third reconstruction unit 1504 , a third measurement unit 1506 , a third determination unit 1508 and a first recommendation unit 1510 .
[0236] The third identification unit 1502 is used to identify a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles.
[0237] The third reconstruction unit 1504 is used to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured.
[0238] The third measuring unit 1506 is used to measure at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0239] The third determining unit 1508 is used to determine the size data of the part model as the predicted size data of the corresponding part.
[0240] The first recommendation unit 1510 is used to recommend product attribute information matching the predicted size data, wherein the clothing products meeting the product attribute information are to be worn by the corresponding parts, and the clothing products are from a virtual clothing store deployed on the e-commerce platform.
[0241] It should be noted that the third identification unit 1502, the third reconstruction unit 1504, the third measurement unit 1506, the third determination unit 1508 and the first recommendation unit 1510 correspond to steps S402 to S410 in Embodiment 1, and the four units and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the above units can be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above units can also be run in the computer terminal 120 provided in Embodiment 3 as part of the device.
[0242] According to an embodiment of the present application, another method for implementing the above Figure 5 The product information recommendation method and the product information recommendation device shown can be applied to a client, and the client is used to deploy a virtual clothing store on an e-commerce platform.
[0243] Fig.16 is a schematic diagram of another device for recommending product information according to an embodiment of the present application, such as Fig.16 As shown, the device 1600 for recommending product information may include: a fourth recognition unit 1602 , a fourth reconstruction unit 1604 , a fourth detection unit 1606 , a fourth determination unit 1608 and a second recommendation unit 1610 .
[0244] The fourth identification unit 1602 is configured to identify a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles.
[0245] The fourth reconstruction unit 1604 is used to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured.
[0246] The fourth detection unit 1606 is used to measure at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0247] The fourth determining unit 1608 is used to determine the size data of the part model as the predicted size data of the corresponding part.
[0248] The second recommendation unit 1610 is used to recommend product attribute information matching the predicted size data to the e-commerce platform, wherein the clothing products meeting the product attribute information are to be worn by the corresponding parts, and the clothing products come from the virtual clothing store.
[0249] It should be noted that the fourth identification unit 1602, the fourth reconstruction unit 1604, the fourth measurement unit 1606, the fourth determination unit 1608 and the second recommendation unit 1610 correspond to steps S502 to S210 in Embodiment 1, and the five units are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the above Embodiment 1. It should be noted that the above units can be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above units can also be run in the computer terminal 120 provided in Embodiment 3 as part of the device.
[0250] According to an embodiment of the present application, another method for implementing the above Figure 6 The human body size measuring method shown is a human body size measuring device.
[0251] Fig.17 is a schematic diagram of another human body size measuring device according to an embodiment of the present application, such as Fig.17 As shown, the human body size measuring device 1700 may include: a first display unit 1702 , a second display unit 1704 and a third display unit 1706 .
[0252] The first display unit 1702 is used to respond to an image input operation on the operation interface and display a plurality of images to be processed on the operation interface, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives.
[0253] The second display unit 1704 is used to respond to the reconstruction operation on the operation interface and display a human body model of the human body to be measured on the operation interface, wherein the human body model is reconstructed from the human body to be measured in the processed image and is used to simulate the geometric contour of the human body to be measured.
[0254] The third display unit 1706 is used to respond to the size measurement operation performed on the operation interface and display the predicted size data of at least one part of the human body to be measured on the operation interface, wherein the predicted size data of the part is determined based on the size data of the corresponding part model in the human body model, and the part model is used to simulate the geometric contour of the corresponding part.
[0255] It should be noted that the first display unit 1702, the second display unit 1704 and the third display unit 1706 correspond to steps S602 to S606 in Embodiment 1, and the three units and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the above units may be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above units may also be part of the device and run in the computer terminal 120 provided in Embodiment 3.
[0256] According to an embodiment of the present application, another method for implementing the above Figure 7 The human body size measuring method shown is a human body size measuring device.
[0257] Fig.18 is a schematic diagram of another human body size measuring device according to an embodiment of the present application, such as Fig.18 As shown, the human body dimension measuring device 1800 may include: a display unit 1802 , a fifth reconstruction unit 1804 , a fifth measurement unit 1806 and a fifth determination unit 1808 .
[0258] The display unit 1802 is used to display multiple images to be processed on a presentation screen of a virtual reality VR device or an augmented reality AR device, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives.
[0259] The fifth reconstruction unit 1804 is used to drive the VR device or the AR device to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured.
[0260] The fifth measuring unit 1806 is used to drive the VR device or the AR device to measure at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured.
[0261] The fifth determining unit 1808 is used to determine the size data of the part model as the predicted size data of the corresponding part, and display the predicted size data of the part on the presentation screen of the VR device or the AR device.
[0262] It should be noted that the above-mentioned display unit 1802, the fifth reconstruction unit 1804, the fifth measurement unit 1806 and the fifth determination unit 1808 correspond to steps S702 to S708 in Embodiment 1, and the four units and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Embodiment 1. It should be noted that the above-mentioned units may be hardware components or software components stored in a memory (e.g., memory 1204) and processed by one or more processors (e.g., processors 1202a, 1202b..., 1202n), and the above-mentioned units may also be part of the device and may be run in the computer terminal 120 provided in Embodiment 3.
[0263] In the object size data measuring device of this embodiment, multiple images to be processed can be obtained, for example, a front human body image and a side human body image, and a human body model (which can be a human body geometry mesh) of the human body to be measured is reconstructed using the multiple images to be processed. By measuring the human body model, the size data of the human body to be measured (for example, human body size information) is obtained, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0264] The embodiment of the present application may provide an electronic device, which may be any electronic device in a group of electronic devices. Optionally, in this embodiment, the electronic device may also be replaced by a terminal device such as a mobile terminal.
[0265] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.
[0266] In this embodiment, the electronic device can execute the program code in the method.
[0267] Optionally, Fig.19 is a structural block diagram of an electronic device according to an embodiment of the present application. Fig.19 As shown, the electronic device A may include: one or more (only one is shown in the figure) processors 1902, a memory 1904, a storage controller, and a peripheral interface, wherein the peripheral interface is connected to a radio frequency module, an audio module, and a display.
[0268] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the methods in the above embodiments. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal A via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0269] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: identifying multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; reconstructing the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measuring at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; and determining the size data of the part model as the predicted size data of the corresponding part.
[0270] Optionally, the processor may also execute program code of the following steps: determining at least one geometric element of a part model in a human body model, wherein the geometric element is used to constitute a surface of the part model; and measuring the geometric element of the part model to obtain size data of the part model.
[0271] Optionally, the processor may also execute the following program code: calling a size measurement model to measure geometric elements of the part model to obtain size data of the part model, wherein the size measurement model is trained based on size data samples of human body samples.
[0272] Optionally, the processor may also execute program codes of the following steps: calling a forward propagation network in the size measurement model, measuring geometric elements of the human body model, and obtaining size data of the part model.
[0273] Optionally, the processor may also execute the program code of the following steps: calling a reconstruction model to reconstruct the human body to be measured in the processed image to obtain a human body model of the human body to be measured, wherein the reconstruction model is obtained by training a human body model sample based on a human body sample, and the human body model sample is used to simulate the geometric contour of the human body sample.
[0274] Optionally, the processor may also execute program codes of the following steps: determining the degree of loss of the predicted size data compared to the actual size data of the part; determining a gradient based on the loss and the geometric element, wherein the gradient is used to represent the degree of influence of the geometric element on the loss; propagating the gradient using a back propagation network in the size measurement model; and using the propagated gradient to perform supervised training on the reconstruction model to obtain a trained reconstruction model.
[0275] Optionally, the processor may also execute the program code of the following steps: using a feature extraction module in the reconstruction model to extract initial image features of the human body to be measured from the image to be processed; using a feature connection module in the reconstruction model to connect multiple initial image features corresponding to multiple images to be processed to obtain target image features; using a regression processing module in the reconstruction model to perform regression processing on the target image features to obtain geometric parameters of the human body to be measured, and generating a human body model based on the geometric parameters.
[0276] Optionally, the processor may also execute program codes of the following steps: determining multiple vertices of a geometric contour of a human body to be measured based on geometric parameters; and creating a three-dimensional geometric mesh from the vertices, wherein the human body model includes the three-dimensional geometric mesh.
[0277] Optionally, the processor may also execute program code of the following steps: deviating the vertices of the reference geometric contour according to the geometric parameters to obtain the vertices of the geometric contour of the human body to be measured.
[0278] Optionally, the processor may also execute the program code of the following steps: identifying a front image to be processed and a side image to be processed, wherein the human body to be measured is displayed in the front image to be processed from a front perspective, and the human body to be measured is displayed in the side image to be processed from a side perspective.
[0279] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: identifying multiple images to be processed, wherein the image contents of the multiple images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; reconstructing the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measuring at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determining the size data of the part model as the predicted size data of the corresponding part; and recommending product attribute information matching the predicted size data, wherein the clothing products meeting the product attribute information are to be worn by the corresponding part, and the clothing products come from a virtual clothing store deployed on the e-commerce platform.
[0280] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: identify multiple images to be processed, wherein the image content of the multiple images to be processed includes the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; reconstruct the human body to be measured in the images to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determine the size data of the part model as the predicted size data of the corresponding part; recommend product attribute information matching the predicted size data to the e-commerce platform, wherein the clothing products meeting the product attribute information are to be worn by the corresponding part, and the clothing products come from the virtual clothing store.
[0281] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: in response to an image input operation on the operation interface, display a plurality of images to be processed on the operation interface, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; in response to a reconstruction operation on the operation interface, display a human body model of the human body to be measured on the operation interface, wherein the human body model is obtained by reconstructing the human body to be measured in the image to be processed and is used to simulate the geometric contour of the human body to be measured; in response to a size measurement operation on the operation interface, display predicted size data of at least one part of the human body to be measured on the operation interface, wherein the predicted size data of the part is determined based on the size data of the corresponding part model in the human body model, and the part model is used to simulate the geometric contour of the corresponding part.
[0282] The processor can call the information and application program stored in the memory through the transmission device to perform the following steps: display multiple images to be processed on the presentation screen of the virtual reality VR device or the augmented reality AR device, wherein the image content of the multiple images to be processed includes the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; drive the VR device or the AR device to reconstruct the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; drive the VR device or the AR device to measure at least one part model in the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part in the human body to be measured; determine the size data of the part model as the predicted size data of the corresponding part, and display the predicted size data of the part on the presentation screen of the VR device or the AR device.
[0283] In the embodiment of the present application, a plurality of images to be processed may be obtained, for example, a front human body image and a side human body image, and a human body model (which may be a human body geometry mesh) of the human body to be measured may be reconstructed using the plurality of images to be processed. By measuring the human body model, the size data of the human body to be measured (for example, human body size information) may be obtained, thereby achieving the technical effect of improving the efficiency of detecting the size data of the human body and solving the technical problem of the efficiency of measuring the size data of the object.
[0284] Those skilled in the art will understand that Fig.19 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, and a mobile Internet device (MID for short). The figure does not limit the structure of the above electronic device. For example, the electronic device A may also include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a configuration different from that shown in the figure.
[0285] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.
[0286] The embodiment of the present application also provides a computer-readable storage medium. Optionally, in this embodiment, the computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the storage medium is located is controlled to execute the program code executed by the method provided in the first embodiment, and the specific execution process is as shown above, which will not be repeated here.
[0287] Optionally, in this embodiment, the computer-readable storage medium may be located in any electronic device in a group of electronic devices in a computer network, or in any mobile terminal in a group of mobile terminals.
[0288] The embodiment of the present application further provides a computer program product. Optionally, in this embodiment, the computer program product may include a computer program, and the computer program implements the method provided in the embodiment when executed by a processor.
[0289] The embodiments of the present application also provide a computer program product. Optionally, the computer program product may include a non-volatile computer-readable storage medium, which may be used to store a computer program, and when the computer program is executed by a processor, the method provided in the embodiments is implemented.
[0290] The embodiment of the present application further provides a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, the method provided in the above embodiment is implemented.
[0291] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0292] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.
[0293] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0294] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0295] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0296] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), mobile hard disk, disk or optical disk and other media that can store program codes.
[0297] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, it does not depart from the principles of the present application.
Claims
1. A method for measuring the size of a human body, characterized in that: include: Identifying a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; Reconstructing the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; Measuring at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured; The size data of the part model is determined as the corresponding predicted size data of the part.
2. The method according to claim 1, characterized in that Measuring at least one part model of the human body model to obtain size data of the part model includes: Determine at least one geometric element of the part model in the human body model, wherein the geometric element is used to constitute a surface of the part model; The geometric elements of the part model are measured to obtain size data of the part model.
3. The method according to claim 2, characterized in that Measuring the geometric elements of the part model to obtain size data of the part model includes: The size measurement model is called to measure the geometric elements of the part model to obtain the size data of the part model, wherein the size measurement model is obtained by training based on size data samples of human body samples.
4. The method according to claim 3, characterized in that Calling a size measurement model to measure the geometric elements of the human body model to obtain size data of the part model includes: The forward propagation network in the size measurement model is called to measure the geometric elements of the human body model to obtain the size data of the part model.
5. The method according to claim 3, characterized in that: Reconstructing the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, including: The reconstruction model is called to reconstruct the human body to be measured in the image to be processed to obtain the human body model of the human body to be measured, wherein the reconstruction model is obtained by training a human body model sample based on the human body sample, and the human body model sample is used to simulate the geometric contour of the human body sample.
6. The method according to claim 5, characterized in that The method further comprises: Determining a degree of loss of the predicted size data compared to actual size data of the part; Determining a gradient based on the loss degree and the geometric element, wherein the gradient is used to represent the degree of influence of the geometric element on the loss degree; Propagate the gradient using a back propagation network in the size measurement model; The propagated gradient is used to perform supervised training on the reconstruction model to obtain the trained reconstruction model.
7. The method according to claim 5, characterized in that Calling a reconstruction model to reconstruct the human body to be measured in the image to be processed to obtain the human body model of the human body to be measured, including: Using the feature extraction module in the reconstruction model, extracting the initial image features of the human body to be measured from the image to be processed; Using the feature connection module in the reconstruction model, multiple initial image features corresponding to multiple images to be processed are connected to obtain target image features; The target image features are subjected to regression processing using a regression processing module in the reconstruction model to obtain geometric parameters of the human body to be measured, and the human body model is generated based on the geometric parameters.
8. The method according to claim 7, characterized in that Generating the human body model based on the geometric parameters includes: Based on the geometric parameters, determining a plurality of vertices of the geometric profile of the human body to be measured; The vertices are created as a three-dimensional geometric mesh, wherein the human body model includes the three-dimensional geometric mesh.
9. The method according to claim 8, characterized in that Based on the geometric parameters, a plurality of vertices of the geometric profile of the human body to be measured are determined, including: According to the geometric parameters, the vertices of the reference geometric contour are deviated to obtain the vertices of the geometric contour of the human body to be measured.
10. The method according to claim 3, characterized in that: The human body samples include human body samples wearing clothing products of different sizes.
11. The method according to any one of claims 1 to 10, characterized in that Multiple images to be processed are identified, including: A front image to be processed and a side image to be processed are identified, wherein the human body to be measured is displayed in the front image to be processed from a front perspective, and the human body to be measured is displayed in the side image to be processed from a side perspective.
12. A method for recommending product information, characterized in that: Applied to e-commerce platforms, including: Identifying a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; Reconstructing the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; Measuring at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured; Determining the size data of the part model as the corresponding predicted size data of the part; Product attribute information matching the predicted size data is recommended, wherein the clothing product satisfying the product attribute information is to be worn by the corresponding part, and the clothing product comes from a virtual clothing store deployed on the e-commerce platform.
13. A method for recommending product information, characterized in that: Applied to a client, the client is used to deploy a virtual clothing store on an e-commerce platform, and the method includes: Identifying a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; Reconstructing the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; Measuring at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured; Determining the size data of the part model as the corresponding predicted size data of the part; Product attribute information matching the predicted size data is recommended to the e-commerce platform, wherein the clothing products satisfying the product attribute information are to be worn by the corresponding parts, and the clothing products are from the virtual clothing store.
14. A method for measuring the size of a human body, characterized in that: include: In response to an image input operation on the operation interface, a plurality of images to be processed are displayed on the operation interface, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; In response to a reconstruction operation performed on the operation interface, a human body model of the human body to be measured is displayed on the operation interface, wherein the human body model is obtained by reconstructing the human body to be measured in the image to be processed and is used to simulate the geometric contour of the human body to be measured; In response to a size measurement operation performed on the operation interface, predicted size data of at least one part of the human body to be measured is displayed on the operation interface, wherein the predicted size data of the part is determined based on size data of a corresponding part model in the human body model, and the part model is used to simulate a geometric contour of the corresponding part.
15. A method for measuring the size of a human body, characterized in that: include: On a presentation screen of a virtual reality (VR) device or an augmented reality (AR) device, a plurality of images to be processed are displayed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed from different perspectives; Driving the VR device or the AR device to reconstruct the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; Driving the VR device or the AR device to measure at least one part model of the human body model to obtain size data of the part model, wherein the part model is used to simulate the geometric contour of the corresponding part of the human body to be measured; The size data of the part model is determined as the corresponding predicted size data of the part, and the predicted size data of the part is displayed on a presentation screen of the VR device or the AR device.
16. A human body size measuring device, characterized in that: include: A first recognition unit is used to recognize a plurality of images to be processed, wherein the image contents of the plurality of images to be processed include the same human body to be measured, and the human body to be measured is displayed in different images to be processed at different viewing angles; A first reconstruction unit is used to reconstruct the human body to be measured in the image to be processed to obtain a human body model of the human body to be measured, wherein the human body model is used to simulate the geometric contour of the human body to be measured; A first measuring unit, used for measuring at least one part model of the human body model to obtain size data of the part model, wherein the part model is used for simulating the geometric contour of the corresponding part of the human body to be measured; The first determining unit is used to determine the size data of the part model as the corresponding predicted size data of the part.
Citation Information
Patent Citations
Artificial neural network-based multi-source gait feature extraction and identification method
CN101807245A
Human body measurement method based on motion reconstruction
CN111915725A
Non-contact infant height measuring method, device and equipment and storage medium
CN117179744A
Method for generating virtual image, and storage medium
WO2024153010A1
