A virtual fitting method, device, system and electronic equipment
By acquiring images of the target object from different angles and constructing a 3D image using key points of the human body and auxiliary imaging points, the problem of high difficulty and low accuracy in 3D reconstruction in virtual try-on is solved, achieving more efficient 3D imaging and improved user experience.
Patent Information
- Application Number
- CN202010761493.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-07-31
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-07-31
AI Technical Summary
Existing virtual try-on technology suffers from low 3D imaging accuracy due to the low resolution of depth cameras and the difficulty of 3D reconstruction caused by human movement.
By acquiring multiple images of the target object from different angles, a three-dimensional image is constructed using human body key point information and auxiliary imaging point information, and then synthesized with a clothing sample library to display the synthesized fitting image.
It enhances the 3D reconstruction performance and 3D imaging accuracy under human motion conditions, thereby improving the user experience.
Smart Images

Figure CN112070575B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of virtual fitting, and in particular to a virtual fitting method, device, system and electronic equipment. Background Art
[0002] With the rapid development of China's clothing e-commerce, online clothing sales have grown rapidly, and its sales proportion in the entire clothing sales has become higher and higher. Compared with the traditional clothing sales model, online clothing sales have many incomparable advantages. On the one hand, there are a wide variety of clothes in the mall, and it is often inconvenient to try them on one by one and make a choice, so it is not easy to find clothes that suit you. On the other hand, the fitting rooms in the mall are often limited, especially during holidays, there are often long queues at the door of the fitting rooms, which affects the buyers' desire to try on clothes and their purchase ideas. Therefore, fast and realistic virtual fitting technology is about to emerge, and has gradually become a research and development hotspot for enterprises and research institutions.
[0003] The current virtual fitting technology mostly uses a monocular depth camera for video acquisition. The monocular depth camera can provide information on four channels of the Red Green Blue-Depth Map (RGB-D), namely: the three colors of red, blue, and green, as well as the distance between the object surface and the camera. Each pixel corresponds to a point in the real world. After obtaining the RGB-D acquisition information, the three-dimensional imaging of the human body is completed through the three-dimensional reconstruction technology, and the virtual fitting operation is further completed.
[0004] However, due to the low resolution of current depth cameras and the occlusion between object surfaces, the data provided by depth cameras is relatively sparse. In addition, during the fitting process, the human body will swing, rotate, and perform other movements, which will further increase the difficulty of 3D reconstruction and reduce the accuracy of 3D imaging. Summary of the invention
[0005] The purpose of the embodiments of the present application is to provide a virtual fitting method, device, system and electronic device, which can solve the problem of difficulty in three-dimensional reconstruction and low accuracy of three-dimensional imaging caused by human body movement in the existing fitting process.
[0006] In a first aspect of the present invention, a virtual fitting method is provided, the method comprising:
[0007] Acquire a plurality of first images of the target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information;
[0008] constructing a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image;
[0009] The three-dimensional image is synthesized with a first target clothing sample in a clothing sample library, and a fitting image after the synthesis is displayed.
[0010] Optionally, acquiring a plurality of first images of the target object at different angles includes:
[0011] Acquire multiple first videos at different angles, where the first videos are compressed and encoded videos;
[0012] The first video is decompressed to obtain a decompressed second video, wherein the first image is a frame image in the second video.
[0013] Optionally, after acquiring a plurality of first images of the target object at different angles, the method further includes:
[0014] The motion posture of the target object is acquired according to the human body key point information and the auxiliary imaging point information in the first image.
[0015] Optionally, synthesizing the three-dimensional image with a first target clothing sample in a clothing sample library and displaying the synthesized fitting image includes:
[0016] According to the three-dimensional image, identifying the human body contour of the target object;
[0017] Processing the first target clothing sample to obtain a processed second target clothing sample;
[0018] Virtually stitching the human body contour and the second target clothing sample to obtain a stitched image after virtual stitching;
[0019] The fitting image is acquired and displayed according to the stitched image and the motion posture.
[0020] Optionally, the human body key point information includes: position information of the human body key point parts of the target object.
[0021] Optionally, the auxiliary imaging point information includes: position information of a part of the target object that is at a preset distance from the key point of the human body.
[0022] In a second aspect of the present invention, a virtual fitting device is provided, comprising:
[0023] A first acquisition module, used to acquire a plurality of first images of a target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information;
[0024] A first processing module, configured to construct a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image;
[0025] The second processing module is used to synthesize the three-dimensional image with a first target clothing sample in the clothing sample library, and display the fitting image after the synthesis processing.
[0026] In a third aspect of the present invention, a virtual fitting system is provided, comprising: an image acquisition system, an image transmission system, an image processing system and a clothing sample library; the image acquisition system is connected to the image processing system via the image transmission system, and the clothing sample library is connected to the image processing system;
[0027] Among them, the image acquisition system acquires multiple first videos of the target object at different angles, and sends the first videos to the image transmission system, the image transmission system compresses and encodes the first videos, and sends the compressed and encoded first videos to the image processing system, the image processing system constructs a three-dimensional image of the target object based on the human body key point information and auxiliary imaging point information in the first image in the first video, and synthesizes the three-dimensional image with the first target clothing sample in the clothing sample library, and displays the synthesized fitting image.
[0028] In the fourth aspect of the implementation of the present invention, there is also provided an electronic device, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the virtual fitting method as described above.
[0029] In a fifth aspect of the present invention, a computer-readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the virtual fitting method described above are implemented.
[0030] Compared with the prior art, the present invention has the following advantages:
[0031] In an embodiment of the present invention, by acquiring multiple first images of a target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information, a three-dimensional image of the target object is constructed based on the human body key point information and the auxiliary imaging point information, and the three-dimensional image is synthesized with a first target clothing sample in a clothing sample library, and the synthesized fitting image is displayed, which can not only enhance the three-dimensional reconstruction performance of the human body in motion, but also improve the accuracy of three-dimensional imaging, thereby further improving the user experience.
[0032] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0034] Figure 1 A block diagram of a virtual fitting system provided by an embodiment of the present invention;
[0035] Figure 2 A block diagram of an image acquisition system provided by an embodiment of the present invention;
[0036] Figure 3 A schematic flow chart of a virtual fitting method provided by an embodiment of the present invention;
[0037] Figure 4 A block diagram of a virtual fitting device provided by an embodiment of the present invention;
[0038] Figure 5 A block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0039] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are 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 are within the scope of protection of this application.
[0040] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.
[0041] like Figure 1 As shown, an embodiment of the present invention provides a virtual fitting system, comprising: an image acquisition system 11, an image transmission system 12, an image processing system 13 and a clothing sample library 14; the image acquisition system 11 is connected to the image processing system 13 through the image transmission system 12, and the clothing sample library 14 is connected to the image processing system 13;
[0042] Among them, the image acquisition system 11 acquires multiple first videos of the target object at different angles, and sends the first video to the image transmission system 12, the image transmission system 12 compresses and encodes the first video, and sends the compressed and encoded first video to the image processing system 13, the image processing system 13 constructs a three-dimensional image of the target object according to the human body key point information and auxiliary imaging point information in the first image in the first video, and synthesizes the three-dimensional image with the first target clothing sample in the clothing sample library 14, and displays the synthesized fitting image.
[0043] Specifically, Figure 2 As shown, the image acquisition system 11 may include a main camera and an auxiliary camera, the main camera may include a first camera 111, and the auxiliary camera may include a second camera 112 and a third camera 113. The first camera 111 may be a high-definition single-lens depth camera, which is responsible for collecting the front image of the target object, and can mainly obtain the main data such as facial expressions and body contours. The high-definition image taken by the first camera is used for overall imaging restoration, face recognition, and human body contour detail extraction. In addition, the auxiliary camera (such as the second camera 112 and the third camera 113) is placed at different positions on the side of the target object with a certain inclination. The auxiliary camera can be an infrared camera with relatively low accuracy. Its main purpose is to obtain the three-dimensional contour of the human body at different angles, so as to better reconstruct the three-dimensional human body image and the virtual fitting image. The video data of the main camera and the auxiliary camera can be collected in a synchronous manner and sent to the image transmission system 12 (the image transmission system 12 is a multi-channel video transmission system). The image transmission system 12 can compress and encode the multi-channel first video through the compression video encoding technology and transmit it to the subsequent image processing system 13 (the image processing system 13 is a multi-channel video processing system), which can reduce the transmission bandwidth. The compression coding may be compression coding based on Discrete Cosine Transform (DCT) or compression coding based on Discrete Wavelet Transform (DWT), which is not limited here.
[0044] It should be noted that the number, position, angle, etc. of the above-mentioned main cameras and auxiliary cameras are only examples and are not specifically limited here.
[0045] Specifically, Figure 3 As shown, an embodiment of the present invention provides a virtual fitting method, which is applied to an image processing system 13 and includes:
[0046] Step 301: Acquire a plurality of first images of a target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information.
[0047] Specifically, firstly, a plurality of first images of the target object are acquired, and the plurality of first images may be images at different angles. Each of the first images contains not only the human body key point information of the target object, but also the auxiliary imaging point information for assisting the imaging of the target object.
[0048] Optionally, the human body key point information may include: position information of the human body key point parts of the target object.
[0049] Specifically, the key points of the human body may include characteristic parts of the human body such as the head, neck, and shoulders. The position information of the key points of the human body is the coordinate position of each key point of the human body.
[0050] Optionally, the auxiliary imaging point information includes: position information of a part of the target object that is at a preset distance from the key point of the human body.
[0051] Specifically, the target object is provided with a plurality of auxiliary imaging points, and the positions of the auxiliary imaging points can be around the key points of the human body, that is, the coordinate positions separated by a preset distance from the coordinate positions of the key points of the human body. The auxiliary imaging points are provided around the key points of the human body, and can be used to assist in identifying the precise contours of the key points of the human body, and further identify the external contours of the entire human body. The auxiliary imaging points can also be provided at other positions other than the key points of the human body, and used to assist in identifying the contours of the human body at other positions of the target object other than the key points of the human body.
[0052] It should be noted that the preset distance can be a specific distance value or a distance range value, which is used to determine whether the position meets the distance limit value as an auxiliary imaging point. If the preset distance is a distance range value, all parts of the target object that are within the preset distance from the key point of the human body can be set as auxiliary imaging points.
[0053] Among them, the auxiliary imaging points can be positioned by means of patches, setting light-emitting diodes (LEDs), etc., which can further enhance the effect of data splicing and restoration of human body contour details. In the subsequent three-dimensional human body reconstruction, the details of the human body contour can be identified with high precision by identifying the auxiliary imaging points.
[0054] Step 302: construct a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image.
[0055] Specifically, for data stitching and fusion of multiple images, matching points are generally obtained based on surface features or are manually marked. Manual marking requires human intervention and cannot achieve real-time stitching, so it is only applicable to the static test stage. Therefore, it is necessary to stitch images by means of matching points obtained through surface features. In addition, for three-dimensional reconstruction and virtual fitting, the recognition and matching of the details of the human body contour also directly determines the effect and user experience of the virtual fitting system. The embodiment of the present invention uses a feature electrical matching algorithm for data stitching and three-dimensional reconstruction, which can further enhance the performance of three-dimensional reconstruction and virtual fitting.
[0056] Here, the key points of the human body and the auxiliary imaging points can be used as anchor points, and the position information of the anchor points can be used as prior information. The matching error is calculated and iterated to continuously reduce the matching error, so as to achieve a better matching effect. Among them, the algorithm flow based on the continuous iteration of the feature point matching algorithm is as follows: the input data is two point cloud data sets containing the source point set P and the target point set Q. The feature point matching algorithm searches for the corresponding matching point pairs, calculates the transformation matrix, unifies the coordinate system of the source point and the target point, calculates the error function of the two point clouds after the transformation, and then determines whether the error function meets the condition of being less than the preset threshold. If the condition is not met, it is iterated again until the error function is less than the preset threshold.
[0057] Step 303: synthesize the three-dimensional image with the first target clothing sample in the clothing sample library, and display the synthesized fitting image.
[0058] Specifically, the clothing sample library contains multiple clothing samples, and a clothing sample to be tried on is selected and determined as a first clothing sample from the multiple clothing samples. By synthesizing the three-dimensional image and the first target clothing sample, a synthetic fitting image is obtained, and the fitting image is displayed for the user to view the fitting effect.
[0059] In the above embodiment of the present invention, by acquiring multiple first images of different angles about the target object, the first images contain human body key point information and auxiliary imaging point information, and constructing a three-dimensional image about the target object based on the human body key point information and auxiliary imaging point information, and synthesizing the three-dimensional image with the first target clothing sample in the clothing sample library, and displaying the synthesized fitting image, not only can the three-dimensional reconstruction performance of the human body in motion be enhanced, but also the accuracy of three-dimensional imaging can be improved, thereby further improving the user experience.
[0060] Optionally, step 301 may specifically include:
[0061] Acquire multiple first videos at different angles, where the first videos are compressed and encoded videos;
[0062] The first video is decompressed to obtain a decompressed second video, wherein the first image is a frame image in the second video.
[0063] Specifically, firstly, a first video after compression encoding is received, wherein the number of the first videos is multiple, and each of the first videos is a video shot at different angles. Then, the first video is decompressed to obtain a decompressed second video, wherein each of the second videos includes multiple frame images, and the first image is a frame image in the second video.
[0064] Optionally, after step 301, the method may further include:
[0065] The motion posture of the target object is acquired according to the human body key point information and the auxiliary imaging point information in the first image.
[0066] Specifically, according to different three-dimensional human body reconstruction algorithms, intermediate data in the process of three-dimensional human body reconstruction using the first image may be used as parameters for acquiring the motion posture of the target object.
[0067] For example, if the method of 3D human reconstruction adopts a feature point matching algorithm, the processing flow of 3D human reconstruction is as follows: 3D human reconstruction is performed based on the human key point information and auxiliary imaging point information in the first image. In the process of 3D human reconstruction, intermediate data, i.e., the feature point data of the image, can be obtained; based on the feature point data, the motion posture of the target object can be obtained. Moreover, based on the feature point data, the 3D human reconstruction process can be continued to obtain the 3D image of the target object, which can further reduce the complexity of the virtual fitting process.
[0068] Optionally, after acquiring the first image of the target object, three-dimensional human body reconstruction can be performed based on the human body key point information and auxiliary imaging point information in the first image, so as to construct a three-dimensional image of the target object. The motion posture of the target object can be directly obtained based on the human body key point information and auxiliary imaging point information in the first image, such as: the motion trend of the target object can be obtained based on the human body key point information and auxiliary imaging point information in a plurality of consecutive first images, so as to obtain the motion posture of the target object. Alternatively, the motion posture of the target object can be obtained based on the human body key point information and auxiliary imaging point information in the three-dimensional image. Alternatively, the motion posture of the target object can be obtained based on the human body key point information and auxiliary imaging point information in the three-dimensional image and the human body key point information and auxiliary imaging point information in the first image, which is not specifically limited here.
[0069] Optionally, the step 303 may specifically include:
[0070] According to the three-dimensional image, identifying the human body contour of the target object;
[0071] Processing the first target clothing sample to obtain a processed second target clothing sample;
[0072] Virtually stitching the human body contour and the second target clothing sample to obtain a stitched image after virtual stitching;
[0073] The fitting image is acquired and displayed according to the stitched image and the motion posture.
[0074] Specifically, according to the human key point information and auxiliary imaging point information in the three-dimensional image, the human body contour of the target object can be identified. In other words, it is necessary to automatically identify the human body contour from the three-dimensional image through a human body image recognition algorithm in order to perform subsequent virtual stitching. Then, the first target clothing sample to be tried on is determined from the clothing sample library, and the first target clothing sample is processed (such as clothing deformation processing, clothing effect simulation processing, etc.) through a clothing material construction algorithm to obtain a processed second target clothing sample; then the human body contour and the second target clothing sample are virtually stitched together to obtain a stitched image after virtual stitching, and according to the stitched image and the motion posture, a fitting image can be obtained, and the fitting image is output for the user to watch the fitting effect.
[0075] It should be noted that the movement posture may include the posture angle of the human body, the movement direction of the human body (or the direction of the movement trend of the human body, the direction of the human body turning angle), etc., which are not specifically limited here.
[0076] To sum up, in the embodiments of the present invention, by acquiring first images of the target object from different multiple angles, a three-dimensional image of the target object is constructed according to the human body key point information and the auxiliary imaging point information, and the three-dimensional image is synthesized with the first target clothing sample in the clothing sample library, and the synthesized fitting image is displayed, which can not only enhance the three-dimensional reconstruction performance and speed in the human body motion state, but also improve the accuracy of three-dimensional imaging and the matching accuracy of virtual stitching, thereby further improving the user experience.
[0077] like Figure 4 As shown, a virtual fitting device 400 provided by an embodiment of the present invention includes:
[0078] A first acquisition module 401 is used to acquire a plurality of first images of a target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information;
[0079] A first processing module 402, configured to construct a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image;
[0080] The second processing module 403 is used to synthesize the three-dimensional image with the first target clothing sample in the clothing sample library, and display the synthesized fitting image.
[0081] In the above embodiment of the present invention, a plurality of first images of different angles about a target object are acquired through a first acquisition module 401, wherein the first images contain human key point information and auxiliary imaging point information, and a first processing module 402 constructs a three-dimensional image about the target object according to the human key point information and the auxiliary imaging point information, and a second processing module 403 synthesizes the three-dimensional image with a first target clothing sample in a clothing sample library, and displays the synthesized fitting image, which can not only enhance the three-dimensional reconstruction performance under human motion state, but also improve the accuracy of three-dimensional imaging, thereby further improving the user experience.
[0082] Optionally, the first acquisition module 401 includes:
[0083] A first acquisition unit, configured to acquire a plurality of first videos at different angles, wherein the first videos are compressed and encoded videos;
[0084] The second acquisition unit is used to decompress the first video to obtain a decompressed second video, wherein the first image is a frame image in the second video.
[0085] Optionally, the device further comprises:
[0086] The second acquisition module is used to acquire the motion posture of the target object according to the human body key point information and the auxiliary imaging point information in the first image.
[0087] Optionally, the second processing module 403 includes:
[0088] A recognition unit, used to recognize the human body contour of the target object according to the three-dimensional image;
[0089] a third acquisition unit, configured to process the first target clothing sample to acquire a processed second target clothing sample;
[0090] a fourth acquisition unit, configured to virtually sew the human body contour and the second target clothing sample to acquire a sewed image after the virtual sew;
[0091] A fifth acquisition unit is used to acquire and display the fitting image according to the stitched image and the motion posture.
[0092] Optionally, the human body key point information includes: position information of the human body key point parts of the target object.
[0093] Optionally, the auxiliary imaging point information includes: position information of a part of the target object that is at a preset distance from the key point of the human body.
[0094] To sum up, in the embodiments of the present invention, by acquiring first images of the target object from different multiple angles, a three-dimensional image of the target object is constructed according to the human body key point information and the auxiliary imaging point information, and the three-dimensional image is synthesized with the first target clothing sample in the clothing sample library, and the synthesized fitting image is displayed, which can not only enhance the three-dimensional reconstruction performance and speed in the human body motion state, but also improve the accuracy of three-dimensional imaging and the matching accuracy of virtual stitching, thereby further improving the user experience.
[0095] It should be noted that the embodiment of the virtual fitting device is a device corresponding to the above-mentioned virtual fitting method. All implementation methods of the above-mentioned method embodiment are applicable to the embodiment of the device and can achieve the same technical effects, which will not be repeated here.
[0096] The embodiment of the present invention further provides an electronic device, such as Figure 5 As shown, it includes a processor 501 , a communication interface 502 , a memory 503 and a communication bus 504 , wherein the processor 501 , the communication interface 502 , and the memory 503 communicate with each other via the communication bus 504 .
[0097] The memory 503 is used to store computer programs.
[0098] When the electronic device is a terminal device, the processor 501 is used to execute the program stored in the memory 503 to implement the following steps:
[0099] Acquire a plurality of first images of the target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information;
[0100] constructing a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image;
[0101] The three-dimensional image is synthesized with a first target clothing sample in a clothing sample library, and a fitting image after the synthesis is displayed.
[0102] Optionally, when the processor 501 is used to execute the above program stored in the memory 503, the following steps are specifically implemented:
[0103] Acquire multiple first videos at different angles, where the first videos are compressed and encoded videos;
[0104] The first video is decompressed to obtain a decompressed second video, wherein the first image is a frame image in the second video.
[0105] Optionally, after acquiring a plurality of first images of the target object at different angles, the processor 501 further implements the following steps when used to execute a program stored in the memory 503:
[0106] The motion posture of the target object is acquired according to the human body key point information and the auxiliary imaging point information in the first image.
[0107] Optionally, when the processor 501 is used to execute the above program stored in the memory 503, the following steps are specifically implemented:
[0108] According to the three-dimensional image, identifying the human body contour of the target object;
[0109] Processing the first target clothing sample to obtain a processed second target clothing sample;
[0110] Virtually stitching the human body contour and the second target clothing sample to obtain a stitched image after virtual stitching;
[0111] The fitting image is acquired and displayed according to the stitched image and the motion posture.
[0112] Optionally, the human body key point information includes: position information of the human body key point parts of the target object.
[0113] Optionally, the auxiliary imaging point information includes: position information of a part of the target object that is at a preset distance from the key point of the human body.
[0114] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
[0115] The communication interface is used for communication between the above terminal and other devices.
[0116] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.
[0117] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0118] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions. When the instructions are executed on a computer, the computer executes the virtual fitting method described in the above embodiment.
[0119] In another embodiment of the present invention, a computer program product including instructions is provided. When the computer program product is run on a computer, the computer executes the virtual fitting method described in the above embodiment.
[0120] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.
[0121] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0122] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims
1. A virtual fitting method, characterized in that: include: Acquire a plurality of first images of the target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information; constructing a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image; Combining the three-dimensional image with a first target clothing sample in a clothing sample library, and displaying a fitting image after the synthesis process; After acquiring a plurality of first images at different angles of the target object, the method comprises: The three-dimensional image of the target object is constructed by using a feature point matching algorithm; feature point data of the image is obtained according to the human body key point information and auxiliary imaging point information in the first image; and the motion posture of the target object is obtained according to the feature point data; The step of synthesizing the three-dimensional image with a first target clothing sample in a clothing sample library and displaying a fitting image after the synthesis process comprises: According to the three-dimensional image, identifying the human body contour of the target object; Processing the first target clothing sample to obtain a processed second target clothing sample; Virtually stitching the human body contour and the second target clothing sample to obtain a stitched image after virtual stitching; Acquire and display the fitting image according to the stitched image and the motion posture; Among them, the motion posture includes the posture angle of the human body, the movement direction of the human body, the direction of the human body's movement trend and the direction of the human body's turning angle; the human body key point information includes the position information of the human body key point parts of the target object; the auxiliary imaging point information includes the position information of the parts on the target object that are spaced a preset distance from the human body key point parts.
2. The method according to claim 1, characterized in that The step of acquiring a plurality of first images of the target object at different angles comprises: Acquire multiple first videos at different angles, where the first videos are compressed and encoded videos; The first video is decompressed to obtain a decompressed second video, wherein the first image is a frame image in the second video.
3. A virtual fitting device, characterized in that: include: A first acquisition module, used to acquire a plurality of first images of a target object at different angles, wherein the first images contain human body key point information and auxiliary imaging point information; A first processing module, configured to construct a three-dimensional image of the target object according to the human body key point information and the auxiliary imaging point information in the first image; A second processing module is used to synthesize the three-dimensional image with a first target clothing sample in the clothing sample library, and display a fitting image after the synthesis process; A second acquisition module, used for acquiring the motion posture of the target object according to the human body key point information and the auxiliary imaging point information in the first image; Specifically used to construct the three-dimensional image of the target object using a feature point matching algorithm; obtain feature point data of the image according to the human body key point information and auxiliary imaging point information in the first image; Acquiring the motion posture of the target object according to the feature point data; The second processing module comprises: A recognition unit, used to recognize the human body contour of the target object according to the three-dimensional image; a third acquisition unit, configured to process the first target clothing sample to acquire a processed second target clothing sample; a fourth acquisition unit, configured to virtually sew the human body contour and the second target clothing sample to acquire a sewed image after the virtual sew; a fifth acquiring unit, configured to acquire and display the fitting image according to the stitched image and the motion posture; Among them, the motion posture includes the posture angle of the human body, the movement direction of the human body, the direction of the human body's movement trend and the direction of the human body's turning angle; the human body key point information includes the position information of the human body key point parts of the target object; the auxiliary imaging point information includes the position information of the parts on the target object that are spaced a preset distance from the human body key point parts.
4. A virtual fitting system, characterized in that: include: Image acquisition system, image transmission system, image processing system and clothing sample library; The image acquisition system is connected to the image processing system through the image transmission system, and the clothing sample library is connected to the image processing system; The image acquisition system acquires a plurality of first videos of the target object at different angles, and sends the first videos to the image transmission system; the image transmission system compresses and encodes the first videos, and sends the compressed and encoded first videos to the image processing system; the image processing system constructs a three-dimensional image of the target object according to the human body key point information and auxiliary imaging point information in the first image in the first video, and synthesizes the three-dimensional image with the first target clothing sample in the clothing sample library, and displays the fitting image after the synthesis processing; After obtaining the human body key point information and the auxiliary imaging point information in the first image in the first video, the method further includes: The three-dimensional image of the target object is constructed by using a feature point matching algorithm; feature point data of the image is obtained according to the human body key point information and auxiliary imaging point information in the first image; and the motion posture of the target object is obtained according to the feature point data; The three-dimensional image is synthesized with the first target clothing sample in the clothing sample library to display a fitting image after the synthesis, including: According to the three-dimensional image, identifying the human body contour of the target object; Processing the first target clothing sample to obtain a processed second target clothing sample; Virtually stitching the human body contour and the second target clothing sample to obtain a stitched image after virtual stitching; Acquire and display the fitting image according to the stitched image and the motion posture; Among them, the motion posture includes the posture angle of the human body, the movement direction of the human body, the direction of the human body's movement trend and the direction of the human body's turning angle; the human body key point information includes the position information of the human body key point parts of the target object; the auxiliary imaging point information includes the position information of the parts on the target object that are spaced a preset distance from the human body key point parts.
5. An electronic device, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the virtual fitting method as described in any one of claims 1 to 2.
6. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, the steps of the virtual fitting method according to any one of claims 1 to 2 are implemented.
