A face 3D modeling method, device and equipment and storage medium

Abstract

The present application relates to the technical field of 3D modeling, in particular to a face 3D modeling method, device and equipment and storage medium. The present application first establishes a preset track of an image acquisition device. Since the preset track is set around the sampled user, the image acquisition device can be controlled to move along the preset track around the sampled user, thereby collecting face images of each angle. Only when face images of each angle are collected, the 3D face model of the sampled user can be completely restored. From the above analysis, the present application is that the image acquisition device moves around the sampled user instead of the sampled user rotating to make the image acquisition device obtain images of each angle. The present application adopts the former, on the one hand, the face images of each angle can be accurately obtained by controlling the image acquisition device, on the other hand, the moving frequency of the sampled user can be reduced, thereby increasing the experience of the sampled user.

Description

Technical Field

[0001] This invention relates to the field of 3D modeling technology, specifically to a method, apparatus, device, and storage medium for 3D face modeling. Background Technology

[0002] Equipment solutions for 3D face modeling can be categorized into 2D simulation and 3D structured light / TOF combined with a 2D camera to achieve 3D face modeling. While pure 2D camera solutions can convert 2D models using prior knowledge, they suffer from a serious drawback: poor accuracy.

[0003] Existing 3D structured light / TOF+2D camera solutions fall into two extremes. One extreme is large-scale equipment, which uses multiple cameras distributed around the face to simultaneously capture facial image data. This offers high accuracy but is also costly, time-consuming, and unsuitable for mass production. The other extreme is low-cost, but because the cameras are fixed, capturing complete data requires the user to turn their head left or right as instructed. This is limited by various uncontrollable factors during facial rotation, making high-precision 3D facial data reconstruction impossible and resulting in low customer acceptance.

[0004] In summary, the accuracy of 3D face modeling in existing technologies is relatively low.

[0005] Therefore, existing technologies still need to be improved and enhanced. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method, apparatus, device, and storage medium for 3D face modeling, which solves the problem of low fidelity in existing 3D face modeling.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a 3D face modeling method, wherein the modeling method includes:

[0009] A preset trajectory is planned for the image acquisition device, the preset trajectory surrounding the user being sampled;

[0010] The image acquisition device is controlled to move along the preset trajectory to acquire several facial images of the sampled user from various angles;

[0011] A 3D face model is constructed based on several of the aforementioned face images.

[0012] In one implementation, the preset trajectory of the planned image acquisition device, the preset trajectory surrounding the sampled user, includes:

[0013] Collect the historical movement trajectory of the sampled user prior to being sampled;

[0014] Based on the historical movement trajectory, the predicted movement trajectory of the sampled user is obtained;

[0015] Based on the predicted movement trajectory of the sampled user, a preset trajectory for the image acquisition device is planned.

[0016] In one implementation, the preset trajectory of the planned image acquisition device, the preset trajectory surrounding the sampled user, includes:

[0017] A predicted trajectory is determined around the sampled user, and the predicted trajectory is a loop centered on the sampled user;

[0018] Mark a first point on the predicted trajectory, the first point pointing radially toward the center of the sampled user;

[0019] Mark the second point and the third point on both sides of the first point. The angle between the line connecting the second point and the sampled user and the line connecting the third point and the sampled user is greater than 90 degrees.

[0020] A ring is drawn based on the first point, the second point, and the third point to plan the preset trajectory of the image acquisition device.

[0021] In one implementation, controlling the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles includes:

[0022] Monitor the real-time location of the sampled user;

[0023] Calculate the real-time distance between the real-time position and the initial image acquisition device, wherein the initial image acquisition device is located at the first point of the preset trajectory;

[0024] When the real-time distance is less than the preset distance, the image acquisition device is controlled to move along the preset trajectory to acquire several facial images of the sampled user from various angles.

[0025] In one implementation, when the real-time distance is less than a preset distance, controlling the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles includes:

[0026] When the real-time distance is less than the focusing distance of the image acquisition device in the preset distance, the image acquisition device is controlled to move from the first point of the preset trajectory, and then repeatedly move to the second point, the first point, and the third point.

[0027] The image acquisition device, which is controlled to move repeatedly, acquires facial images of the sampled user at the first point, the second point, and the third point, respectively, thereby obtaining several facial images of the sampled user from various angles.

[0028] In one implementation, constructing a 3D face model based on a plurality of the face images includes:

[0029] Remove overlapping face images captured by the image acquisition device at the same location on the preset trajectory to obtain the preprocessed face image;

[0030] Based on the preprocessed face image, the images acquired by the image acquisition device at the first point, the second point, and the third point are obtained and denoted as the first image, the second image, and the third image, respectively.

[0031] Calculate the grayscale value of each pixel in the second image and the third image;

[0032] Remove pixels with gray values ​​greater than a gray threshold to obtain a second segmented image corresponding to the second image and a third segmented image corresponding to the third image.

[0033] A 3D face model is constructed based on the first image, the second segmented image, and the third segmented image.

[0034] In one implementation, the modeling method further includes a modeling device, which comprises the following components:

[0035] A distance sensor is mounted on the image acquisition device;

[0036] The main control unit has its input terminal electrically connected to the output terminal of the distance sensor;

[0037] The drive unit has its input terminal electrically connected to the output terminal of the main control unit and its mechanical connection to the image acquisition device.

[0038] Secondly, embodiments of the present invention also provide a 3D face modeling device, wherein the device comprises the following components:

[0039] A trajectory planning module is used to plan a preset trajectory for the image acquisition device, the preset trajectory surrounding the user being sampled;

[0040] The control module is used to control the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles;

[0041] The modeling module is used to construct a 3D face model based on several of the aforementioned face images.

[0042] Thirdly, embodiments of the present invention also provide a terminal device, wherein the terminal device includes a memory, a processor, and a face 3D modeling program stored in the memory and executable on the processor, wherein when the processor executes the face 3D modeling program, it implements the steps of the face 3D modeling method described above.

[0043] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a face 3D modeling program, wherein when the face 3D modeling program is executed by a processor, it implements the steps of the face 3D modeling method described above.

[0044] Beneficial Effects: This invention first establishes a preset trajectory for the image acquisition device. Since this preset trajectory is set around the user being sampled, the image acquisition device can be controlled to move around the user along the preset trajectory, thereby acquiring facial images from various angles. Only by acquiring facial images from various angles can the 3D facial model of the user be completely reconstructed. From the above analysis, it can be seen that this invention uses the movement of the image acquisition device around the user, rather than the user rotating, to acquire images from various angles. This former approach allows for precise acquisition of facial images from various angles by controlling the image acquisition device, while also reducing the frequency of movement for the user, thus enhancing their user experience. Attached Figure Description

[0045] Figure 1 This is an overall flowchart of the present invention;

[0046] Figure 2 This is a schematic diagram of the preset trajectory coverage area in an embodiment of the present invention;

[0047] Figure 3 This is a schematic diagram of the preset trajectory path in an embodiment of the present invention;

[0048] Figure 4 This is a system diagram of the modeling equipment in an embodiment of the present invention;

[0049] Figure 5 This is a flowchart of the 3D modeling process in an embodiment of the present invention;

[0050] Figure 6 This is a block diagram illustrating the internal structure of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0052] Research has revealed that 3D face modeling equipment solutions can be categorized into 2D simulation and a combination of 3D structured light / TOF + 2D cameras to achieve 3D face modeling. While pure 2D camera solutions can convert 2D models using prior knowledge, they suffer from serious accuracy issues. Existing 3D structured light / TOF + 2D camera solutions fall into two extremes: one is large-scale equipment that uses multiple cameras distributed around the face to simultaneously capture facial image data, offering high accuracy but also high cost and long modeling time, hindering mass production; the other is low-cost, but because the cameras are fixed, capturing complete data requires the user to turn their head left or right as instructed, limiting the accuracy of 3D face data reconstruction due to various uncontrollable factors during facial rotation, resulting in low customer acceptance.

[0053] To address the aforementioned technical problems, this invention provides a method, apparatus, device, and storage medium for 3D face modeling, resolving the issue of low fidelity in existing 3D face modeling. Specifically, the invention first plans a preset trajectory for the image acquisition device, then controls the image acquisition device to move along the preset trajectory to achieve the purpose of moving the image acquisition device around the sampled user, thereby acquiring several face images of the sampled user from various angles; finally, based on these face images, a 3D face model is constructed.

[0054] For example, constructing a 3D face model requires five facial images of the user from different angles. Therefore, this embodiment first plans a preset trajectory for the camera (image acquisition device). As long as the camera moves around this preset trajectory, it can acquire five facial images of the user from different angles. These five facial images from different angles can then be used to reconstruct the user's 3D face model. Since this embodiment does not require user movement, but only requires precise control of the camera's movement to acquire facial images from specified angles, the accuracy of the acquired facial images is improved, thereby enhancing the fidelity of the 3D model.

[0055] Exemplary methods

[0056] The 3D face modeling method of this embodiment can be applied to terminal devices, which can be terminal products with image acquisition capabilities, such as computers and mobile phones. In this embodiment, as... Figure 1 As shown, the 3D face modeling method specifically includes the following steps:

[0057] S100, Plan the preset trajectory of the image acquisition device, the preset trajectory surrounding the user being sampled.

[0058] In this embodiment, the image acquisition device is a camera. First, a preset trajectory for the camera is planned, and then the camera is controlled to move along the preset trajectory to acquire the user's facial image. In this embodiment, planning the preset trajectory includes two methods. The first method is for users who are moving. Based on the user's historical movement trajectory, their possible movement trajectory is predicted, and then the preset trajectory of the camera is planned based on the user's predicted trajectory, allowing the camera to move around the user to acquire facial images from various angles. This method is suitable for users who are moving. The second method is for users with a very small or almost no movement range. In this case, only the preset range of the camera needs to be planned based on the user's position. This method can acquire facial images from various angles of the user and reduces the computational load due to the smaller amount of positional data acquired. When step S100 is the first method, step S100 includes the following steps S101, S102, and S103:

[0059] S101, Collect the historical movement trajectory of the sampled user before the sampling.

[0060] S102, based on the historical movement trajectory, obtain the predicted movement trajectory of the sampled user.

[0061] In this embodiment, predicting a user's future movement trajectory based on their historical movement is a prior art technique.

[0062] S103, based on the predicted movement trajectory of the sampled user, plan the preset trajectory of the image acquisition device.

[0063] When step S100 is the second case, step S100 includes the following steps S104, S105, S106, and S107:

[0064] S104, determine a predicted trajectory around the sampled user, the predicted trajectory being a loop centered on the sampled user.

[0065] In this embodiment, a rough estimated trajectory is first determined around the sampled user, with the sampled user as the center of the estimated trajectory. The reason for establishing the estimated trajectory first is to facilitate the gradual acquisition of a precise preset trajectory, so that the preset trajectory can cover all positions on the sampled user's face.

[0066] S105, mark a first point on the estimated trajectory, the first point pointing radially toward the center of the sampled user along the estimated trajectory.

[0067] The first point is to ensure that the camera can capture images when it is located at that position. Figure 2 The center of the user's face is shown.

[0068] S106, mark the second point and the third point on both sides of the first point, and the angle between the line connecting the second point and the sampled user and the line connecting the third point and the sampled user is greater than 90 degrees.

[0069] The second and third points are respectively Figure 2 The left and right sides of the face. An angle greater than 90 degrees is to ensure that the camera can cover the entire area of ​​the face when moving along the preset trajectory.

[0070] S107, Draw a ring based on the first point, the second point and the third point, and plan the preset trajectory of the image acquisition device.

[0071] This embodiment uses three points to draw a ring, which is a prior art technique.

[0072] S200, control the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles.

[0073] In this embodiment, in addition to the first point, second point, and third point mentioned above, the preset trajectory also includes points such as... Figure 3 The points shown, among which Figure 3 In the diagram, "1" represents the first point, "2" represents the second point, and "4" represents the fourth point. Within one shooting cycle, the image acquisition device moves and captures images along the prescribed path 1-2-3-2-1-4-5-4-1. During this movement along the predetermined arc-shaped track, at least three 2D images and point cloud photos are captured at each fixed position. As shown in the diagram, five positions (equally divided within a 100-degree range) are captured, for a total of fifteen images, ensuring the integrity of the face data. Similarly, the number of positions is not limited to five, nor is it limited to three photos per position; a comprehensive evaluation based on the system's shooting and image processing capabilities is required. Based on the above principle, step S200 includes the following steps: S201, S202, S203, and S204:

[0074] S201, Monitor the real-time location of the sampled user.

[0075] S202, calculate the real-time distance between the real-time position and the initial image acquisition device, wherein the initial image acquisition device is located at the first point of the preset trajectory.

[0076] S203, when the real-time distance is less than the focusing distance of the image acquisition device in the preset distance, control the image acquisition device to start moving from the first point of the preset trajectory, and repeatedly move to the second point, the first point, and the third point in sequence.

[0077] In this embodiment, the image acquisition device is placed at the first point before acquiring an image. Figure 3 At position "1" in the image acquisition device, when the distance sensor on the image acquisition device detects that the distance to the sampled user is less than 450mm, the image acquisition device begins to capture images and starts to move along... Figure 3 Move along the path 1-2-3-2-1-4-5-4-1.

[0078] S204, the image acquisition device, which is repeatedly moving, acquires the face image of the sampled user at the first point, the second point, and the third point, respectively, to obtain several face images of the sampled user from various angles.

[0079] In this embodiment, the image acquisition device stops to acquire a face image every time it moves to a position point. That is, the image acquisition device in this embodiment acquires images in a stationary state, which can ensure that the acquired images are clear enough.

[0080] S300, construct a 3D face model based on several of the aforementioned face images.

[0081] This embodiment first preprocesses several acquired facial images, and then constructs a 3D facial model. Constructing a 3D facial model is an existing technology. Step S300 includes the following steps:

[0082] S301, Remove overlapping face images acquired by the image acquisition device at the same location on the preset trajectory to obtain the preprocessed face image.

[0083] Removing overlapping images can reduce computational cost, thereby improving modeling speed.

[0084] S302, based on the preprocessed face image, the images acquired by the image acquisition device at the first point, the second point and the third point are obtained, and are respectively denoted as the first image, the second image and the third image.

[0085] S303, Calculate the grayscale value of each pixel in the second image and the third image.

[0086] S304, remove pixels with gray values ​​greater than the gray threshold to obtain a second segmented image corresponding to the second image and a third segmented image corresponding to the third image.

[0087] Depend on Figure 3 It can be seen that the second point ( Figure 3 The "2" in the middle) and the third point ( Figure 3The image captured in “4” is a side view image of a face, which includes hair. This part does not need to be modeled, so it needs to be removed. Since the grayscale value of black hair is very large, this embodiment sets the grayscale threshold to 256 to remove black hair.

[0088] S305, construct a 3D face model based on the first image, the second segmented image, and the third segmented image.

[0089] The first image, the second segmented image, and the third segmented image cover the entire face, thus enabling a highly accurate reconstruction of the 3D face model.

[0090] In one embodiment, a modeling device is also provided, such as Figure 4 As shown, the modeling equipment includes the following components:

[0091] A distance sensor is mounted on the image acquisition device.

[0092] The main control unit (main control SOC system unit) has its input terminal electrically connected to the output terminal of the distance sensor.

[0093] The drive unit has its input terminal electrically connected to the output terminal of the main control unit and its mechanical connection to the image acquisition device.

[0094] The screen display unit has its input terminal electrically connected to the output terminal of the main control unit.

[0095] by Figure 5 The detailed process of 3D modeling in this invention is illustrated by the following example:

[0096] The equipment's arc track has a range of 120 degrees (the arc of the preset circular track is 120 degrees) to ensure that the angles at shooting points 3 and 5 are within 100 degrees (i.e., Figure 3 The arc corresponding to "3" and "5" is 100 degrees, and the distance between 3 and 5 is 700mm (arc length). The distance between the face and the center shooting position 1 is 450mm. That is to say, the face data sampling unit as a whole needs to move and shoot according to the prescribed path 1-2-3-2-1-4-5-4-1 within one shooting cycle. During the movement along the predetermined arc track, at least 3 2D images and point cloud photos are taken at each fixed position. As shown in the figure, there are 5 positions (equally divided within a 100-degree range), and a total of 15 images are taken to ensure the integrity of the face data. Similarly, it is not limited to 5 positions, nor is it limited to taking 3 photos at each position. It needs to be comprehensively evaluated based on the system's shooting capabilities and image processing capabilities.

[0097] In summary, this invention first establishes a preset trajectory for the image acquisition device. Since this preset trajectory is set around the user being sampled, the image acquisition device can be controlled to move around the user along the preset trajectory, thereby acquiring facial images from various angles. Only by acquiring facial images from each angle can the 3D facial model of the user be completely reconstructed. From the above analysis, it can be seen that this invention uses the movement of the image acquisition device around the user, rather than the user rotating, to acquire images from various angles. This former approach allows for precise acquisition of facial images from various angles by controlling the image acquisition device, and also reduces the frequency of movement for the user being sampled, thus enhancing their user experience.

[0098] In addition, the design of this invention does not require users to perform operations such as head shaking. The 2D / 3D camera module is moved by a motor (drive unit) to ensure the accuracy of sampling and modeling, which greatly improves the user experience.

[0099] Exemplary device

[0100] This embodiment also provides a 3D face modeling device, which includes the following components:

[0101] A trajectory planning module is used to plan a preset trajectory for the image acquisition device, the preset trajectory surrounding the user being sampled;

[0102] The control module is used to control the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles;

[0103] The modeling module is used to construct a 3D face model based on several of the aforementioned face images.

[0104] Based on the above embodiments, the present invention also provides a terminal device, the principle block diagram of which can be as follows: Figure 6 As shown, the terminal device includes a processor, memory, network interface, display screen, and temperature sensor connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a 3D modeling method. The display screen can be an LCD screen or an e-ink screen. The temperature sensor is pre-installed inside the terminal device to detect the operating temperature of the internal components.

[0105] Those skilled in the art will understand that Figure 6 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal device to which the present invention is applied. The specific terminal device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0106] In one embodiment, a terminal device is provided, comprising a memory, a processor, and a 3D modeling program stored in the memory and executable on the processor. When the processor executes the 3D modeling program, it implements the following operation instructions:

[0107] A preset trajectory is planned for the image acquisition device, the preset trajectory surrounding the user being sampled;

[0108] The image acquisition device is controlled to move along the preset trajectory to acquire several facial images of the sampled user from various angles;

[0109] A 3D face model is constructed based on several of the aforementioned face images.

[0110] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for 3D face modeling, characterized in that, The modeling method includes: A preset trajectory is planned for the image acquisition device, the preset trajectory surrounding the user being sampled; The image acquisition device is controlled to move along the preset trajectory to acquire several facial images of the sampled user from various angles; Based on the aforementioned facial images, a 3D facial model is constructed; The preset trajectory surrounds the sampled user and includes: A predicted trajectory is determined around the sampled user, and the predicted trajectory is a loop centered on the sampled user; Mark a first point on the predicted trajectory, the first point pointing radially toward the center of the sampled user; Mark the second point and the third point on both sides of the first point. The angle between the line connecting the second point and the sampled user and the line connecting the third point and the sampled user is greater than 90 degrees. A ring is drawn based on the first point, the second point, and the third point to plan the preset trajectory of the image acquisition device; The control of the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles includes: Monitor the real-time location of the sampled user; Calculate the real-time distance between the real-time position and the initial image acquisition device, wherein the initial image acquisition device is located at the first point of the preset trajectory; When the real-time distance is less than the preset distance, and when the real-time distance is less than the focusing distance of the image acquisition device in the preset distance, the image acquisition device is controlled to move from the first point of the preset trajectory, and then move to the second point, the first point, and the third point in sequence. The image acquisition device, which is controlled to move repeatedly, acquires the face image of the sampled user at the positions of the first point, the second point, and the third point, respectively, to obtain several face images of the sampled user from various angles. The step of constructing a 3D face model based on several face images includes: Remove overlapping face images captured by the image acquisition device at the same location on the preset trajectory to obtain the preprocessed face image; Based on the preprocessed face image, the images acquired by the image acquisition device at the first point, the second point, and the third point are obtained and denoted as the first image, the second image, and the third image, respectively. Calculate the grayscale value of each pixel in the second image and the third image; Remove pixels with gray values ​​greater than a gray threshold to obtain a second segmented image corresponding to the second image and a third segmented image corresponding to the third image. A 3D face model is constructed based on the first image, the second segmented image, and the third segmented image.

2. The 3D face modeling method as described in claim 1, characterized in that, The preset trajectory of the planned image acquisition device, the preset trajectory surrounding the user being sampled, includes: Collect the historical movement trajectory of the sampled user prior to being sampled; Based on the historical movement trajectory, the predicted movement trajectory of the sampled user is obtained; Based on the predicted movement trajectory of the sampled user, a preset trajectory for the image acquisition device is planned.

3. The 3D face modeling method as described in claim 1, characterized in that, The modeling method further includes a modeling device, which comprises the following components: A distance sensor is mounted on the image acquisition device; The main control unit has its input terminal electrically connected to the output terminal of the distance sensor; The drive unit has its input terminal electrically connected to the output terminal of the main control unit and its mechanical connection to the image acquisition device.

4. A 3D face modeling device, characterized in that, The device comprises the following components: A trajectory planning module is used to plan a preset trajectory for the image acquisition device, the preset trajectory surrounding the user being sampled; The control module is used to control the image acquisition device to move along the preset trajectory to acquire images. The sampled user has several facial images from various angles. The modeling module is used to construct a 3D face model based on several of the aforementioned face images; The preset trajectory surrounds the sampled user and includes: A predicted trajectory is determined around the sampled user, and the predicted trajectory is a loop centered on the sampled user; Mark a first point on the predicted trajectory, the first point pointing radially toward the center of the sampled user; Mark the second point and the third point on both sides of the first point. The angle between the line connecting the second point and the sampled user and the line connecting the third point and the sampled user is greater than 90 degrees. A ring is drawn based on the first point, the second point, and the third point to plan the preset trajectory of the image acquisition device; The control of the image acquisition device to move along the preset trajectory to acquire several facial images of the sampled user from various angles includes: Monitor the real-time location of the sampled user; Calculate the real-time distance between the real-time position and the initial image acquisition device, wherein the initial image acquisition device is located at the first point of the preset trajectory; When the real-time distance is less than the preset distance, and when the real-time distance is less than the focusing distance of the image acquisition device in the preset distance, the image acquisition device is controlled to move from the first point of the preset trajectory, and then move to the second point, the first point, and the third point in sequence. The image acquisition device, which is controlled to move repeatedly, acquires the face image of the sampled user at the positions of the first point, the second point, and the third point, respectively, to obtain several face images of the sampled user from various angles. The step of constructing a 3D face model based on several face images includes: Remove overlapping face images captured by the image acquisition device at the same location on the preset trajectory to obtain the preprocessed face image; Based on the preprocessed face image, the images acquired by the image acquisition device at the first point, the second point, and the third point are obtained and denoted as the first image, the second image, and the third image, respectively. Calculate the grayscale value of each pixel in the second image and the third image; Remove pixels with gray values ​​greater than a gray threshold to obtain a second segmented image corresponding to the second image and a third segmented image corresponding to the third image. A 3D face model is constructed based on the first image, the second segmented image, and the third segmented image.

5. A terminal device, characterized in that, The terminal device includes a memory, a processor, and a face 3D modeling program stored in the memory and executable on the processor. When the processor executes the face 3D modeling program, it implements the steps of the face 3D modeling method as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a face 3D modeling program, which, when executed by a processor, implements the steps of the face 3D modeling method as described in any one of claims 1-3.

Images