Virtual trial assembly method, device, system and computer-readable storage medium

Abstract

The present invention discloses a virtual fitting method, device, system and computer-readable storage medium. The method comprises: acquiring a two-dimensional video, calculating the coordinates of key points of a character in a current frame of the two-dimensional video, and calculating the character rotation, character scaling and character displacement of the current frame according to the coordinates of the key points of the character and a preset clothing model; calculating rendering parameters according to the character rotation, character scaling and character displacement, and rendering the preset clothing model onto the character in the current frame according to the rendering parameters; the present invention only needs to process the two-dimensional video, greatly reducing the amount of calculation and improving the real-time performance of the virtual fitting; calculating the rendering parameters according to the calculated character rotation, character scaling and character displacement of the current frame, and rendering the preset clothing model onto the character in the current frame according to the rendering parameters, thereby improving the authenticity of the virtual fitting.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a virtual fitting method, device, system and computer-readable storage medium. Background Art

[0002] With the development of society, people have an increasing demand for virtual fitting. Currently, a 3D human body model can be obtained through technologies such as depth cameras or visual 3D reconstruction, and virtual fitting can be performed based on the 3D human body model. However, this will increase the amount of calculation and reduce real-time performance. Deep learning technology can also be used to adapt and deform 2D clothing graphics according to the results of human body image segmentation, and finally render them onto the human body screen, but this will make the rendering less realistic. Since existing methods generally have shortcomings such as low real-time performance and authenticity, resulting in poor user experience, how to improve the real-time performance and authenticity of virtual fitting is an urgent problem to be solved. Summary of the Invention

[0003] The main purpose of the present invention is to provide a virtual fitting method, device, system and computer-readable storage medium, aiming to solve the problem of how to improve the real-time and authenticity of virtual fitting.

[0004] To achieve the above object, the present invention provides a virtual fitting method, which comprises the following steps:

[0005] Acquire a two-dimensional video, calculate the coordinates of key points of a person in a current frame of the two-dimensional video, and calculate the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the coordinates of the key points of the person and a preset clothing model;

[0006] Rendering parameters are calculated according to the character rotation amount, the character scaling amount, and the character displacement amount, and the preset clothing model is rendered onto the character in the current frame according to the rendering parameters.

[0007] Preferably, the step of acquiring a two-dimensional video and calculating the coordinates of key points of a person in a current frame of the two-dimensional video includes:

[0008] Acquire a two-dimensional video, extract a current frame of the two-dimensional video according to a recording order of the two-dimensional video, and determine whether the current frame is an initial frame of the two-dimensional video;

[0009] If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame;

[0010] If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the character in the current frame are calculated based on the position of the character in the current frame, and the coordinates of the key points of the character in the current frame are calculated based on the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame.

[0011] Preferably, the step of calculating the coordinates of the character key points of the current frame according to the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame includes:

[0012] Calculating a position deviation based on the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame, and comparing the position deviation with a preset deviation threshold to obtain a comparison result;

[0013] If the comparison result shows that the position deviation is less than the preset deviation threshold, determining the coordinates of the key points of the character in the current frame according to the pre-coordinates of the key points of the character in the current frame;

[0014] If the comparison result shows that the position deviation is not less than the preset deviation threshold, the coordinates of the character key points of the current frame are calculated based on the preset weight coefficient, the pre-coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame.

[0015] Preferably, the step of calculating the character rotation amount, character scaling amount, and character displacement amount of the current frame according to the character key point coordinates and the preset clothing model includes:

[0016] Obtaining the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculating the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy;

[0017] Obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character's scaling amount in the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount;

[0018] The character neck coordinates in the character key point coordinates are obtained, and the character displacement of the current frame is calculated according to the character neck coordinates, a preset clothing model, the character rotation amount, and the character scaling amount.

[0019] Preferably, the step of calculating the character rotation amount of the current frame according to the character shoulder coordinates and the adaptive gain strategy includes:

[0020] Calculating the character's rotation angle according to the character's shoulder coordinates, and calculating a calibration coefficient according to the character's rotation angle and an adaptive gain strategy;

[0021] The character rotation angle is calibrated according to the calibration coefficient to calculate the character rotation amount of the current frame.

[0022] Preferably, the step of calculating the character scaling amount of the current frame according to the character shoulder coordinates, the character leg coordinates, the preset clothing model and the character rotation amount includes:

[0023] Obtaining a camera matrix and a projection matrix, and extracting the model shoulder coordinates and the model leg coordinates of a preset clothing model, and calculating a first character model matrix based on the camera matrix, the projection matrix, the character shoulder coordinates, the character leg coordinates, the model shoulder coordinates, and the model leg coordinates;

[0024] A default character displacement is obtained, and a character scaling amount of a current frame is calculated according to the default character displacement, the character first model matrix, and the character rotation amount.

[0025] Preferably, the step of calculating the character displacement of the current frame according to the character neck coordinates, the preset clothing model, the character rotation amount, and the character scaling amount includes:

[0026] Obtaining a camera matrix and a projection matrix, and extracting the model neck coordinates of a preset clothing model, and calculating a second character model matrix based on the camera matrix, the projection matrix, the character neck coordinates, and the model neck coordinates;

[0027] The character displacement of the current frame is calculated according to the character rotation amount, the character scaling amount and the second character model matrix.

[0028] In addition, to achieve the above-mentioned purpose, the present invention further provides a virtual fitting device, the virtual fitting device comprising:

[0029] a calculation module, configured to obtain a two-dimensional video, calculate the coordinates of key points of a person in a current frame of the two-dimensional video, and calculate the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the coordinates of the key points of the person and a preset clothing model;

[0030] A rendering module is used to calculate rendering parameters according to the character rotation amount, the character scaling amount and the character displacement amount, and render the preset clothing model onto the character in the current frame according to the rendering parameters.

[0031] Furthermore, the calculation module is also used for:

[0032] Acquire a two-dimensional video, extract a current frame of the two-dimensional video according to a recording order of the two-dimensional video, and determine whether the current frame is an initial frame of the two-dimensional video;

[0033] If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame;

[0034] If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the character in the current frame are calculated based on the position of the character in the current frame, and the coordinates of the key points of the character in the current frame are calculated based on the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame.

[0035] Furthermore, the calculation module is also used for:

[0036] Calculating a position deviation based on the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame, and comparing the position deviation with a preset deviation threshold to obtain a comparison result;

[0037] If the comparison result shows that the position deviation is less than the preset deviation threshold, determining the coordinates of the key points of the character in the current frame according to the pre-coordinates of the key points of the character in the current frame;

[0038] If the comparison result shows that the position deviation is not less than the preset deviation threshold, the coordinates of the character key points of the current frame are calculated based on the preset weight coefficient, the pre-coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame.

[0039] Furthermore, the calculation module is also used for:

[0040] Obtaining the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculating the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy;

[0041] Obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character's scaling amount in the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount;

[0042] The character neck coordinates in the character key point coordinates are obtained, and the character displacement of the current frame is calculated according to the character neck coordinates, a preset clothing model, the character rotation amount, and the character scaling amount.

[0043] Furthermore, the calculation module is also used for:

[0044] Calculating the character's rotation angle according to the character's shoulder coordinates, and calculating a calibration coefficient according to the character's rotation angle and an adaptive gain strategy;

[0045] The character rotation angle is calibrated according to the calibration coefficient to calculate the character rotation amount of the current frame.

[0046] Furthermore, the calculation module is also used for:

[0047] Obtaining a camera matrix and a projection matrix, and extracting the model shoulder coordinates and the model leg coordinates of a preset clothing model, and calculating a first character model matrix based on the camera matrix, the projection matrix, the character shoulder coordinates, the character leg coordinates, the model shoulder coordinates, and the model leg coordinates;

[0048] A default character displacement is obtained, and a character scaling amount of a current frame is calculated according to the default character displacement, the character first model matrix, and the character rotation amount.

[0049] Furthermore, the calculation module is also used for:

[0050] Obtaining a camera matrix and a projection matrix, and extracting the model neck coordinates of a preset clothing model, and calculating a second character model matrix based on the camera matrix, the projection matrix, the character neck coordinates, and the model neck coordinates;

[0051] The character displacement of the current frame is calculated according to the character rotation amount, the character scaling amount and the second character model matrix.

[0052] In addition, to achieve the above-mentioned purpose, the present invention also provides a virtual fitting system, which includes: a memory, a processor, and a virtual fitting program stored in the memory and executable on the processor. When the virtual fitting program is executed by the processor, the steps of the virtual fitting method described above are implemented.

[0053] In addition, to achieve the above-mentioned purpose, the present invention further provides a computer-readable storage medium, on which a virtual fitting program is stored. When the virtual fitting program is executed by a processor, the steps of the virtual fitting method described above are implemented.

[0054] The virtual fitting method proposed in the present invention obtains a two-dimensional video, calculates the coordinates of the key points of the character in the current frame of the two-dimensional video, and calculates the character rotation, character scaling and character displacement of the current frame based on the coordinates of the key points of the character and a preset clothing model; calculates rendering parameters based on the character rotation, character scaling and character displacement, and renders the preset clothing model onto the character in the current frame based on the rendering parameters; the present invention only needs to process the two-dimensional video, which greatly reduces the amount of calculation and improves the real-time performance of the virtual fitting; calculates rendering parameters based on the calculated character rotation, character scaling and character displacement of the current frame, and renders the preset clothing model onto the character in the current frame based on the rendering parameters, thereby improving the authenticity of the virtual fitting. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1Schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;

[0056] Figure 2 Schematic diagram of the process of the first embodiment of the virtual trial assembly method of the present invention;

[0057] Figure 3 It is a schematic diagram of the key points of the character of the present invention.

[0058] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0059] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] like Figure 1 As shown, Figure 1 It is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0061] The device in the embodiment of the present invention may be a PC or a server device.

[0062] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0063] Those skilled in the art will understand that Figure 1 The device structure shown in the figure does not constitute a limitation of the device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0064] like Figure 1 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module and a virtual trial installation program.

[0065] Among them, the operating system is a program that manages and controls portable storage devices and software resources, and supports the operation of network communication modules, user interface modules, virtual trial installation programs and other programs or software; the network communication module is used to manage and control the network interface 1002; the user interface module is used to manage and control the user interface 1003.

[0066] exist Figure 1 In the storage device shown, the storage device calls the virtual fitting program stored in the memory 1005 through the processor 1001 and executes the operations in each embodiment of the virtual fitting method described below.

[0067] Based on the above hardware structure, an embodiment of a virtual trial installation method of the present invention is proposed.

[0068] Reference Figure 2 , Figure 2 1 is a flow chart of a first embodiment of a virtual trial fitting method according to the present invention, the method comprising:

[0069] Step S10, acquiring a two-dimensional video, calculating the coordinates of key points of a person in a current frame of the two-dimensional video, and calculating the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the key point coordinates of the person and a preset clothing model;

[0070] Step S20 , calculating rendering parameters according to the character rotation amount, the character scaling amount, and the character displacement amount, and rendering the preset clothing model onto the character in the current frame according to the rendering parameters.

[0071] The virtual fitting method of this embodiment is applied to the virtual fitting system of the communication service agency, and the virtual fitting system can be applied to various fitting software or terminal devices. For the convenience of description, the virtual fitting system is taken as an example for explanation. When the virtual fitting system obtains the two-dimensional video recorded by the user in real time, it extracts the current frame of the two-dimensional video according to the recording order of the two-dimensional video, and determines whether the current frame is the initial frame of the two-dimensional video. If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the characters in the current frame are calculated according to the position of the characters in the current frame. If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the characters in the current frame are calculated according to the position of the characters in the current frame, and the coordinates of the key points of the characters in the current frame are calculated according to the pre-coordinates of the key points of the characters in the current frame and the coordinates of the key points of the characters in the previous frame of the current frame. The virtual fitting system obtains the coordinates of the key points of the character in the coordinates of the key points of the character, and calculates the rotation of the character in the current frame based on the coordinates of the shoulders and the adaptive gain strategy; the virtual fitting system obtains the coordinates of the shoulders and legs in the coordinates of the key points of the character, and calculates the scaling of the character in the current frame based on the coordinates of the shoulders, the legs, the preset clothing model and the rotation of the character; the virtual fitting system obtains the coordinates of the neck in the coordinates of the key points of the character, and calculates the displacement of the character in the current frame based on the coordinates of the neck, the preset clothing model, the rotation of the character and the scaling of the character; the virtual fitting system calculates the rendering parameters based on the rotation of the character, the scaling of the character and the displacement of the character, and renders the preset clothing model onto the character in the current frame based on the rendering parameters. It should be noted that the two-dimensional video is a video captured in real time by an ordinary monocular camera.

[0072] The virtual fitting method of this embodiment obtains a two-dimensional video, calculates the coordinates of the key points of the character in the current frame of the two-dimensional video, and calculates the character rotation, character scaling and character displacement of the current frame based on the coordinates of the key points of the character and the preset clothing model; calculates rendering parameters based on the character rotation, character scaling and character displacement, and renders the preset clothing model onto the character in the current frame based on the rendering parameters; the present invention only needs to process the two-dimensional video, which greatly reduces the amount of calculation and improves the real-time performance of the virtual fitting; calculates rendering parameters based on the calculated character rotation, character scaling and character displacement of the current frame, and renders the preset clothing model onto the character in the current frame based on the rendering parameters, thereby improving the authenticity of the virtual fitting.

[0073] The following describes each step in detail:

[0074] Step S10, acquiring a two-dimensional video, calculating the coordinates of key points of a person in a current frame of the two-dimensional video, and calculating the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the key point coordinates of the person and a preset clothing model;

[0075] In this embodiment, when the virtual fitting system obtains the two-dimensional video recorded by the user in real time, it obtains the current frame of the two-dimensional video, calculates the coordinates of the key points of the characters in the current frame of the two-dimensional video, and calculates the character rotation, character scaling and character displacement of the current frame based on the coordinates of the key points of the characters and the preset clothing model; it can be understood that the two-dimensional video is recorded in real time, and when one frame of video is recorded, the virtual fitting system takes this frame of video as the current frame, calculates the coordinates of the key points of the characters corresponding to the characters contained in the current frame, and calculates the character rotation, character scaling and character displacement of the current frame based on the coordinates of the key points of the characters and the preset clothing model; it should be noted that the preset clothing model is set in the virtual fitting system in advance, and the virtual fitting system includes preset clothing models of various styles. After the user selects the corresponding preset clothing model according to his or her own preferences, he or she starts recording the two-dimensional video.

[0076] Specifically, in step S10, the steps of obtaining a two-dimensional video and calculating the coordinates of key points of a person in a current frame of the two-dimensional video include:

[0077] Step a: acquiring a two-dimensional video, extracting a current frame of the two-dimensional video according to the recording order of the two-dimensional video, and determining whether the current frame is an initial frame of the two-dimensional video;

[0078] In this step, the virtual fitting system obtains the two-dimensional video recorded by the user in real time, and takes each frame of the two-dimensional video as the current frame in turn according to the recording order of the two-dimensional video, and identifies whether there is a corresponding previous frame for the current frame. If not, it is determined that the current frame is the initial frame of the two-dimensional video. If so, it is determined that the current frame is not the initial frame of the two-dimensional video.

[0079] Step b, if the current frame is the initial frame of the two-dimensional video, calculating the coordinates of the key points of the character in the current frame according to the position of the character in the current frame;

[0080] In this step, if the virtual fitting system determines that the current frame is the initial frame of the two-dimensional video, it directly calculates the coordinates of the key points of the characters in the current frame based on the positions of the characters in the current frame; for example: Figure 3 As shown, Figure 3 The virtual fitting system locates the character's neck according to the character's position in the current frame ( Figure 3 Point 2) in the figure, establish the coordinate axis with the character's neck as the origin, and determine Figure 3 The two-dimensional coordinate values ​​(x coordinates and y coordinates) corresponding to points 1 to 14 in the image are directly mapped to the three-dimensional space through the fully connected layer with residual connection, and the z coordinates of each key point of the character are estimated. The z coordinate represents the relative position of each key point of the character to the character's neck ( Figure 3The depth distance of point 2) in the current frame is calculated, and then the coordinates of the key points of the character in the current frame are calculated. It should be noted that the number and position of the key points of the character can be set according to the actual situation. Preferably, Figure 3 There are 14 key points of the character in the game, namely the top of the head, neck, shoulders, elbows, wrists, hips, knees and ankles.

[0081] Step c: If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the characters in the current frame are calculated based on the position of the characters in the current frame, and the coordinates of the key points of the characters in the current frame are calculated based on the pre-coordinates of the key points of the characters in the current frame and the coordinates of the key points of the characters in the previous frame of the current frame.

[0082] In this step, if the virtual fitting system determines that the current frame is not the initial frame of the two-dimensional video, it calculates the pre-coordinates of the key points of the character in the current frame according to the position of the character in the current frame, and calculates the coordinates of the key points of the character in the current frame according to the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame. For example, since there will be a certain offset between the key points of the character corresponding to the upper and lower frames of the two-dimensional video, the virtual fitting system first calculates the pre-coordinates of the key points of the character in the current frame according to the position of the character in the current frame, and then obtains the coordinates of the key points of the character in the previous frame of the current frame, calculates the position deviation between the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame, and calibrates the pre-coordinates of the key points of the character in the current frame according to the position deviation to determine the coordinates of the key points of the character in the current frame, so as to improve the accuracy of the coordinates of the key points of the character, and thereby improve the authenticity of the rendering of the virtual fitting.

[0083] Furthermore, the step of calculating the coordinates of the character key points of the current frame according to the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame includes:

[0084] Step c1, calculating a position deviation based on the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame, and comparing the position deviation with a preset deviation threshold to obtain a comparison result;

[0085] In this step, after calculating the pre-coordinates of the key points of the character in the current frame and obtaining the coordinates of the key points of the character in the previous frame, the virtual fitting system calculates the position deviation based on the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame, and compares the position deviation with the preset deviation threshold to obtain a comparison result; such as: Figure 3As shown, the virtual fitting system calculates the pre-coordinates of the key points of the character from point 1 to point 14 in the current frame, and obtains the coordinates of the key points of the character from point 1 to point 14 in the previous frame of the current frame. Then, the position deviation between point 1 of the two frames is calculated according to the coordinates corresponding to point 1 in the two frames, and the position deviation between point 2 of the two frames is calculated according to the coordinates corresponding to point 2 in the two frames, until the position deviation between point 14 of the two frames is calculated. Then, the position deviation of point 1 to point 14 is compared with the preset deviation threshold value to obtain the comparison result.

[0086] Step c2: if the comparison result shows that the position deviation is less than the preset deviation threshold, determining the coordinates of the key points of the character in the current frame according to the predicted coordinates of the key points of the character in the current frame;

[0087] In this step, if the virtual fitting system obtains a comparison result that the position deviation is less than the preset deviation threshold, the coordinates of the key points of the character in the current frame are determined according to the pre-coordinates of the key points of the character in the current frame; it can be understood that, if Figure 3 As shown, assuming that the position deviations of points 1 to 8 are all less than the preset deviation threshold, the virtual fitting system directly uses the pre-coordinates of the key points of the characters corresponding to points 1 to 8 in the current frame as the coordinates of the key points of the characters corresponding to points 1 to 8 in the current frame, and the coordinates of the key points of the characters corresponding to points 9 to 14 in the current frame are determined according to the following step c3; assuming that the position deviations of points 1 to 14 are all less than the preset deviation threshold, the virtual fitting system directly uses the pre-coordinates of the key points of the characters corresponding to points 1 to 14 in the current frame as the coordinates of the key points of the characters corresponding to points 1 to 14 in the current frame.

[0088] Step c3: If the comparison result shows that the position deviation is not less than the preset deviation threshold, the coordinates of the character key points of the current frame are calculated based on the preset weight coefficient, the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame.

[0089] In this step, if the virtual fitting system obtains a comparison result that the position deviation is not less than the preset deviation threshold, the coordinates of the key points of the character in the current frame are calculated based on the preset weight coefficient, the pre-coordinates of the key points of the character in the current frame, and the coordinates of the key points of the character in the previous frame of the current frame; Figure 3As shown, assuming that the position deviations of points 1 to 8 are all less than the preset deviation threshold, and the position deviations of points 9 to 14 are not less than the preset deviation threshold, the coordinates of the key points of the characters corresponding to points 1 to 8 in the current frame are determined according to the above step c2. For points 9 to 14 in the current frame, the virtual fitting system determines that the weight coefficient of the pre-coordinates of the key points of the characters in the current frame is 0.2, and the weight coefficient of the coordinates of the key points of the characters in the previous frame of the current frame is 0.8 according to the preset weight coefficient. The pre-coordinates of the key points of the characters corresponding to point 9 in the current frame are multiplied by the weight coefficient 0.2, and the coordinates of the key points of the characters corresponding to point 9 in the previous frame of the current frame are multiplied by the weight coefficient 0.8. The pre-coordinates of the key points of the characters corresponding to point 9 multiplied by the weight coefficient 0.2 and the coordinates of the key points of the characters corresponding to point 9 multiplied by the weight coefficient 0.8 are added to obtain the coordinates of the key points of the characters corresponding to point 9 in the current frame. The same operation is performed on points 10 to 14, and the coordinates of the key points of the characters corresponding to points 9 to 14 in the current frame are determined respectively.

[0090] Specifically, in step S10, the step of calculating the character rotation amount, character scaling amount, and character displacement amount of the current frame according to the character key point coordinates and the preset clothing model includes:

[0091] Step d, obtaining the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculating the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy;

[0092] In this step, the virtual fitting system obtains the left shoulder coordinates and the right shoulder coordinates of the character in the key point coordinates of the character in the current frame. When it is determined that the x-coordinate of the character's left shoulder coordinate is not less than the x-coordinate of the character's right shoulder coordinate, it means that the character in the current frame is facing away from the camera. At this time, there is no need to perform subsequent steps, and the next frame of the current frame in the two-dimensional video is obtained, and the steps of calculating the key point coordinates of the character in the current frame of the two-dimensional video and subsequent steps are re-executed; when it is determined that the x-coordinate of the character's left shoulder coordinate is less than the x-coordinate of the character's right shoulder coordinate, it means that the character in the current frame is facing the camera. At this time, the virtual fitting system calculates the rotation amount of the character in the current frame based on the character's left shoulder coordinates, the character's right shoulder coordinates and the adaptive gain strategy.

[0093] Specifically, step d includes:

[0094] Step d1, calculating the character's rotation angle based on the character's shoulder coordinates, and calculating a calibration coefficient based on the character's rotation angle and an adaptive gain strategy;

[0095] Step d2: calibrating the character rotation angle according to the calibration coefficient to calculate the character rotation amount of the current frame.

[0096] In steps d1 to d2, the virtual fitting system calculates the character rotation angle of the current frame based on the character's left shoulder coordinates and the character's right shoulder coordinates, and calculates a calibration coefficient based on the character rotation angle and the adaptive gain strategy. The character rotation angle is then calibrated based on the calibration coefficient to calculate the character rotation amount of the current frame. It should be noted that the formula for calculating the character rotation angle of the current frame by the virtual fitting system is:

[0097] (1) a=arctan((p3.z-p6.z) / (p3.x-p6.x))

[0098] (2)

[0099] In formula (1), a is radians, and p3.z is Figure 3 The z coordinate of the key point of the character corresponding to point 3 in p6.z is Figure 3 The z coordinate of the key point of the character corresponding to point 6 in p3.x is Figure 3 The x coordinate of the key point of the character corresponding to point 3 in p6.x is Figure 3 The x-coordinate of the key point of the character corresponding to point 6 in ; In formula (2), b is the character rotation angle, a is the radian calculated by formula (1), and pi is the pi.

[0100] The virtual fitting system calculates the calibration coefficient based on the character rotation angle and the adaptive gain strategy, and then calibrates the character rotation angle based on the calibration coefficient to calculate the character rotation amount of the current frame:

[0101] (3)

[0102] (4) s = pow(r, 0.67)

[0103] (5)

[0104] In formula (3), r is the absolute value, b is the character rotation angle calculated by formula (2), and abs is the function for finding the absolute value; in formula (4), s is the magnification, r is the absolute value calculated by formula (3), and pow is the power function; in formula (5), b on the left side of the equation is the calibrated character rotation angle.

[0105] Since the z coordinates of the key points of the character in the current frame are estimated, there is some deviation between the character rotation angle calculated by the above formulas (1) and (2) and the real rotation angle of the human body. Therefore, when calculating the character rotation angle, the virtual fitting system combines the adaptive gain strategy, i.e., formula (3), formula (4) and formula (5), to calibrate the character rotation angle so that the calibrated character rotation angle is closer to the real rotation angle of the human body. Then, the character rotation amount of the current frame is obtained according to the calibrated character rotation angle, which helps to improve the authenticity of the virtual fitting rendering.

[0106] Step e, obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character scaling amount of the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount;

[0107] In this step, the virtual fitting system obtains the character's shoulder coordinates and character's leg coordinates in the character's key point coordinates. The character's shoulder coordinates and character's leg coordinates are located on the same side of the character, that is, they can be the character's left shoulder coordinates and character's left leg coordinates, or the character's right shoulder coordinates and character's right leg coordinates; the virtual fitting system calculates the character scaling amount of the current frame based on the character's shoulder coordinates, character's leg coordinates, the preset clothing model and the character's rotation amount.

[0108] Furthermore, step e includes:

[0109] Step e1, obtaining a camera matrix and a projection matrix, and extracting the model shoulder coordinates and the model leg coordinates of a preset clothing model, and calculating a first character model matrix based on the camera matrix, the projection matrix, the character shoulder coordinates, the character leg coordinates, the model shoulder coordinates, and the model leg coordinates;

[0110] Step e2: Obtain a default character displacement, and calculate a character scaling amount of a current frame according to the default character displacement, the first character model matrix, and the character rotation amount.

[0111] In step e1 to step e2, the virtual fitting system obtains the camera matrix and projection matrix set in advance, and extracts the model shoulder coordinates and model leg coordinates of the preset clothing model. The model shoulder coordinates and model leg coordinates are located on the same side of the model, that is, they can be the model left shoulder coordinates and model left leg coordinates, or the model right shoulder coordinates and model right leg coordinates; the virtual fitting system calculates the first character model matrix based on the camera matrix, projection matrix, character shoulder coordinates, character leg coordinates, model shoulder coordinates and model leg coordinates; the virtual fitting system obtains the default character displacement, and calculates the character scaling amount of the current frame based on the default character displacement, the character first model matrix and the character rotation amount; for example: the virtual fitting system obtains the character left shoulder coordinates and character left leg coordinates in the character key point coordinates, correspondingly obtains the model left shoulder coordinates and model left leg coordinates of the preset clothing model, and obtains the camera matrix and projection matrix according to the following formula:

[0112] (6) T3=Project*View*Model*O3

[0113] (7) T9=Project*View*Model*O9

[0114] (8) Calculate the first character model matrix, where, in formula (6), formula (7) and formula (8), O3 and O9 are the coordinates of the model left shoulder and the model left leg in the preset clothing model, View is the camera matrix, Project is the projection matrix, T3 and T9 are the calculated intermediate values, both of which are four-dimensional vectors, abs is a function for finding the absolute value, height is the height corresponding to the current frame, T9.y is the second-dimensional coordinate value of the four-dimensional vector T9, T9.w is the fourth-dimensional coordinate value of the four-dimensional vector T9, T3.y is the second-dimensional coordinate value of the four-dimensional vector T3, T3.w is the fourth-dimensional coordinate value of the four-dimensional vector T3, and p3.y is Figure 3 The x coordinate of the key point coordinate of the character (the coordinate of the character's left shoulder) corresponding to point 3 in the figure, p9.y is Figure 3 The x-coordinate of the character key point coordinate (the character's left leg coordinate) corresponding to point 9 in , Model is the first character model matrix.

[0115] After calculating the first character model matrix, the virtual fitting system obtains the default character displacement and calculates the character scaling using the formula Model = Scale*Rotate*Translate, where Model is the first character model matrix, Scale is the matrix representing the character scaling, Rotate is the matrix representing the character rotation, and Translate is the default character displacement (0, 0, 0).

[0116] Step f, obtaining the character's neck coordinates in the character's key point coordinates, and calculating the character's displacement in the current frame based on the character's neck coordinates, a preset clothing model, the character's rotation amount, and the character's scaling amount.

[0117] Specifically, step f includes:

[0118] Step f1, obtaining a camera matrix and a projection matrix, and extracting the model neck coordinates of a preset clothing model, and calculating a second character model matrix based on the camera matrix, the projection matrix, the character neck coordinates, and the model neck coordinates;

[0119] Step f2: Calculate the character displacement of the current frame according to the character rotation amount, the character scaling amount, and the second character model matrix.

[0120] In steps f1 to f2, the virtual fitting system sets the camera matrix and projection matrix in advance, and extracts the model neck coordinates of the preset clothing model. According to the camera matrix, projection matrix, character neck coordinates, and model neck coordinates, the second character model matrix is ​​calculated. The virtual fitting system calculates the character displacement of the current frame based on the character rotation amount, character scaling amount, and second character model matrix calculated in the above steps. For example, since the camera position corresponding to the virtual fitting system is set in advance, the camera position is always kept on the Z axis of the coordinate system. Therefore, only the displacement in the x and y directions needs to be solved to accurately render the clothing position. Therefore, the virtual fitting system uses the camera matrix, projection matrix, character neck coordinates, and model neck coordinates through the following formula:

[0121] (9) T2=Project*View*Model*O2

[0122] (10) Calculate the second character model matrix. In formula (9) and formula (10), O2 is the model neck coordinate in the preset clothing model, View is the camera matrix, Project is the projection matrix, T2 is the calculated intermediate value, all of which are four-dimensional vectors. Height is the height corresponding to the current frame, T2.y is the second-dimensional coordinate value of the four-dimensional vector T2, T2.w is the fourth-dimensional coordinate value of the four-dimensional vector T2, and xy on the left side of the equal sign and p2.xy on the right side of the equation are both Figure 3 The x-coordinate and y-coordinate of the character key point coordinates (character neck coordinates) corresponding to point 2 in , Model is the second character model matrix.

[0123] After calculating the second character model matrix, the virtual fitting system obtains the character rotation and character scaling calculated in the above steps, and calculates the character displacement using the formula Model = Scale * Rotate * Translate, where Model is the second character model matrix, Scale is the matrix representing the character scaling, Rotate is the matrix representing the character rotation, and Translate is the matrix representing the character displacement.

[0124] Step S20 , calculating rendering parameters according to the character rotation amount, the character scaling amount, and the character displacement amount, and rendering the preset clothing model onto the character in the current frame according to the rendering parameters.

[0125] In this embodiment, after calculating the character rotation, character scaling and character displacement of the current frame, the virtual fitting system calculates the character model matrix of the current frame according to the character rotation, character scaling and character displacement of the current frame, and calculates the rendering matrix (MVP matrix) of the current frame according to the character model matrix of the current frame. Both the character model matrix and the rendering matrix are rendering parameters. The virtual fitting system transmits the rendering parameters to OpenGL (an image processing software) so as to render the preset clothing model onto the character of the current frame according to the rendering parameters through OpenGL, and display it to the user to realize virtual fitting. It should be noted that, after calculating the character rotation, character scaling and character displacement of the current frame, the virtual fitting system calculates the character model matrix of the current frame through the formula Model=Scale*Rotate*Translate, wherein Model is the character model matrix of the current frame, Scale is the matrix representing the character scaling, Rotate is the matrix representing the character rotation, and Translate is the matrix representing the character displacement.

[0126] Furthermore, after rendering the preset clothing model onto the character in the current frame of the two-dimensional model, the virtual fitting system detects whether the current frame is the last frame of the two-dimensional video. If the current frame is the last frame of the two-dimensional video, the process stops; if the current frame is not the last frame of the two-dimensional video, the next frame of the current frame is used as the current frame, and the above steps are repeated. Since the processing steps for each frame are the same as the above steps, they will not be repeated here.

[0127] When the virtual fitting system of this embodiment obtains the two-dimensional video recorded by the user in real time, it extracts the current frame of the two-dimensional video according to the recording order of the two-dimensional video, and determines whether the current frame is the initial frame of the two-dimensional video. If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame; if the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame, and the coordinates of the key points of the character in the current frame are calculated according to the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame; the virtual fitting system obtains the coordinates of the shoulders of the character in the key point coordinates, and calculates the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy; the virtual fitting system obtains the coordinates of the shoulders of the character in the key point coordinates and The character's leg coordinates are used to calculate the character scaling of the current frame based on the character's shoulder coordinates, the character's leg coordinates, the preset clothing model and the character's rotation amount; the virtual fitting system obtains the character's neck coordinates in the character's key point coordinates, and calculates the character's displacement of the current frame based on the character's neck coordinates, the preset clothing model, the character's rotation amount and the character's scaling amount; the virtual fitting system calculates the rendering parameters based on the character's rotation amount, the character's scaling amount and the character's displacement amount, and renders the preset clothing model onto the character of the current frame based on the rendering parameters; the present invention only needs to process the two-dimensional video, which greatly reduces the amount of calculation and improves the real-time performance of the virtual fitting. The rendering parameters are calculated based on the calculated character rotation amount, the character scaling amount and the character displacement amount of the current frame, and the preset clothing model is rendered onto the character of the current frame based on the rendering parameters, thereby improving the authenticity of the virtual fitting.

[0128] The present invention also provides a virtual fitting device. The virtual fitting device of the present invention comprises:

[0129] a calculation module, configured to obtain a two-dimensional video, calculate the coordinates of key points of a person in a current frame of the two-dimensional video, and calculate the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the coordinates of the key points of the person and a preset clothing model;

[0130] A rendering module is used to calculate rendering parameters according to the character rotation amount, the character scaling amount and the character displacement amount, and render the preset clothing model onto the character in the current frame according to the rendering parameters.

[0131] Furthermore, the calculation module is also used for:

[0132] Acquire a two-dimensional video, extract a current frame of the two-dimensional video according to a recording order of the two-dimensional video, and determine whether the current frame is an initial frame of the two-dimensional video;

[0133] If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame;

[0134] If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the character in the current frame are calculated based on the position of the character in the current frame, and the coordinates of the key points of the character in the current frame are calculated based on the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame.

[0135] Furthermore, the calculation module is also used for:

[0136] Calculating a position deviation based on the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame, and comparing the position deviation with a preset deviation threshold to obtain a comparison result;

[0137] If the comparison result shows that the position deviation is less than the preset deviation threshold, determining the coordinates of the key points of the character in the current frame according to the pre-coordinates of the key points of the character in the current frame;

[0138] If the comparison result shows that the position deviation is not less than the preset deviation threshold, the coordinates of the character key points of the current frame are calculated based on the preset weight coefficient, the pre-coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame.

[0139] Furthermore, the calculation module is also used for:

[0140] Obtaining the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculating the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy;

[0141] Obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character's scaling amount in the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount;

[0142] The character neck coordinates in the character key point coordinates are obtained, and the character displacement of the current frame is calculated according to the character neck coordinates, a preset clothing model, the character rotation amount, and the character scaling amount.

[0143] Furthermore, the calculation module is also used for:

[0144] Calculating the character's rotation angle according to the character's shoulder coordinates, and calculating a calibration coefficient according to the character's rotation angle and an adaptive gain strategy;

[0145] The character rotation angle is calibrated according to the calibration coefficient to calculate the character rotation amount of the current frame.

[0146] Furthermore, the calculation module is also used for:

[0147] Obtaining a camera matrix and a projection matrix, and extracting the model shoulder coordinates and the model leg coordinates of a preset clothing model, and calculating a first character model matrix based on the camera matrix, the projection matrix, the character shoulder coordinates, the character leg coordinates, the model shoulder coordinates, and the model leg coordinates;

[0148] A default character displacement is obtained, and a character scaling amount of a current frame is calculated according to the default character displacement, the character first model matrix, and the character rotation amount.

[0149] Furthermore, the calculation module is also used for:

[0150] Obtaining a camera matrix and a projection matrix, and extracting the model neck coordinates of a preset clothing model, and calculating a second character model matrix based on the camera matrix, the projection matrix, the character neck coordinates, and the model neck coordinates;

[0151] The character displacement of the current frame is calculated according to the character rotation amount, the character scaling amount and the second character model matrix.

[0152] The invention also provides a virtual trial installation system.

[0153] The virtual try-out system includes a memory, a processor, and a virtual try-out program stored in the memory and executable on the processor. When the virtual try-out program is executed by the processor, the steps of the virtual try-out method described above are implemented.

[0154] The method implemented when the virtual trial installation program running on the processor is executed can refer to the various embodiments of the virtual trial installation method of the present invention, and will not be described in detail here.

[0155] The present invention also provides a computer-readable storage medium.

[0156] The computer-readable storage medium stores a virtual try-out program, which implements the steps of the virtual try-out method described above when executed by the processor.

[0157] The method implemented when the virtual trial installation program running on the processor is executed can refer to the various embodiments of the virtual trial installation method of the present invention, and will not be described in detail here.

[0158] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0159] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0160] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention is essentially 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) as described above, and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0161] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A virtual fitting method, characterized in that: The virtual fitting method comprises the following steps: Acquire a two-dimensional video, calculate the coordinates of key points of a person in a current frame of the two-dimensional video, and calculate the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the coordinates of the key points of the person and a preset clothing model; Calculating rendering parameters according to the character rotation amount, the character scaling amount, and the character displacement amount, and rendering the preset clothing model onto the character in the current frame according to the rendering parameters; The step of calculating the character rotation amount, character scaling amount, and character displacement amount of the current frame according to the character key point coordinates and the preset clothing model includes: Obtaining the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculating the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy; Obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character's scaling amount in the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount; The character neck coordinates in the character key point coordinates are obtained, and the character displacement of the current frame is calculated according to the character neck coordinates, a preset clothing model, the character rotation amount, and the character scaling amount.

2. The virtual fitting method according to claim 1, wherein: The step of obtaining a two-dimensional video and calculating the coordinates of key points of a person in a current frame of the two-dimensional video includes: Acquire a two-dimensional video, extract a current frame of the two-dimensional video according to a recording order of the two-dimensional video, and determine whether the current frame is an initial frame of the two-dimensional video; If the current frame is the initial frame of the two-dimensional video, the coordinates of the key points of the character in the current frame are calculated according to the position of the character in the current frame; If the current frame is not the initial frame of the two-dimensional video, the pre-coordinates of the key points of the character in the current frame are calculated based on the position of the character in the current frame, and the coordinates of the key points of the character in the current frame are calculated based on the pre-coordinates of the key points of the character in the current frame and the coordinates of the key points of the character in the previous frame of the current frame.

3. The virtual fitting method according to claim 2, wherein: The step of calculating the coordinates of the character key points of the current frame according to the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame includes: Calculating a position deviation based on the predicted coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame, and comparing the position deviation with a preset deviation threshold to obtain a comparison result; If the comparison result shows that the position deviation is less than the preset deviation threshold, determining the coordinates of the key points of the character in the current frame according to the pre-coordinates of the key points of the character in the current frame; If the comparison result shows that the position deviation is not less than the preset deviation threshold, the coordinates of the character key points of the current frame are calculated based on the preset weight coefficient, the pre-coordinates of the character key points of the current frame and the coordinates of the character key points of the previous frame of the current frame.

4. The virtual fitting method according to claim 1, wherein: The step of calculating the character rotation amount of the current frame according to the character shoulder coordinates and the adaptive gain strategy includes: Calculating a character rotation angle according to the character's shoulder coordinates, and calculating a calibration coefficient according to the character rotation angle and an adaptive gain strategy; The character rotation angle is calibrated according to the calibration coefficient to calculate the character rotation amount of the current frame.

5. The virtual fitting method according to claim 1, wherein: The step of calculating the character scaling amount of the current frame according to the character shoulder coordinates, the character leg coordinates, the preset clothing model and the character rotation amount includes: Obtaining a camera matrix and a projection matrix, and extracting the model shoulder coordinates and the model leg coordinates of a preset clothing model, and calculating a first character model matrix based on the camera matrix, the projection matrix, the character shoulder coordinates, the character leg coordinates, the model shoulder coordinates, and the model leg coordinates; A default character displacement is obtained, and a character scaling amount of a current frame is calculated according to the default character displacement, the character first model matrix, and the character rotation amount.

6. The virtual fitting method according to claim 1, wherein: The step of calculating the character displacement of the current frame according to the character neck coordinates, the preset clothing model, the character rotation amount, and the character scaling amount includes: Obtaining a camera matrix and a projection matrix, and extracting the model neck coordinates of a preset clothing model, and calculating a second character model matrix based on the camera matrix, the projection matrix, the character neck coordinates, and the model neck coordinates; The character displacement of the current frame is calculated according to the character rotation amount, the character scaling amount and the second character model matrix.

7. A virtual fitting device, characterized in that: The virtual trial installation device comprises: a calculation module, configured to obtain a two-dimensional video, calculate the coordinates of key points of a person in a current frame of the two-dimensional video, and calculate the rotation amount, scaling amount, and displacement amount of the person in the current frame based on the coordinates of the key points of the person and a preset clothing model; a rendering module, configured to calculate rendering parameters according to the character rotation amount, the character scaling amount, and the character displacement amount, and render the preset clothing model onto the character in the current frame according to the rendering parameters; The calculation module is further configured to obtain the coordinates of the shoulders of the character in the coordinates of the key points of the character, and calculate the rotation amount of the character in the current frame according to the coordinates of the shoulders of the character and the adaptive gain strategy; Obtaining the coordinates of the character's shoulders and legs from the character's key point coordinates, and calculating the character's scaling amount in the current frame based on the coordinates of the character's shoulders, the coordinates of the character's legs, a preset clothing model, and the character's rotation amount; The character neck coordinates in the character key point coordinates are obtained, and the character displacement of the current frame is calculated according to the character neck coordinates, a preset clothing model, the character rotation amount, and the character scaling amount.

8. A virtual fitting system, characterized in that: The virtual try-out system includes: a memory, a processor, and a virtual try-out program stored in the memory and executable on the processor. When the virtual try-out program is executed by the processor, the steps of the virtual try-out method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a virtual try-out program, which, when executed by a processor, implements the steps of the virtual try-out method according to any one of claims 1 to 6.

Images