Model control method and system applied to VR head-mounted display interaction and storage medium

By monitoring the eye-moving cursor and gyroscope of the VR headset, the selection and rotation of the model are achieved, and the problem of increasing cost and difficulty of traditional handle control methods is solved, and the operation efficiency and accuracy of virtual and real technologies are improved.

CN119024970AInactive Publication Date: 2024-11-26VSWORK NETWORK TECHNOLOGY CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411489942.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-24
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The traditional VR headset model control method requires the help of a handle, which increases the cost and operation difficulty, and is not conducive to the development of virtual and real technologies.

Method used

The selection and movement of the model is achieved by monitoring the eye-moving cursor of the VR headset, and the rotation of the model is achieved by using a gyroscope, which simplifies the operation process, provides functional buttons and observation cameras to improve control accuracy.

Benefits of technology

It improves the accuracy and efficiency of model selection and movement, reduces operation difficulty, makes the model's rotation control more accurate and convenient, and is suitable for the application of virtual and realistic technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119024970A_ABST
    Figure CN119024970A_ABST
Patent Text Reader

Abstract

The invention discloses a model control method and system applied to VR head-mounted display interaction and a storage medium, and belongs to the technical field of virtuality and reality, and the method comprises the steps: monitoring an eye movement cursor of a VR head-mounted display; after it is monitored that the eye movement cursor collides with the model, a selection instruction of a user is obtained, and the model is selected as a target model; detecting an operation instruction of a user; if the instruction is a moving instruction, acquiring position coordinates of an eye movement cursor in real time, and moving the target model to the position coordinates after receiving a confirmation message; if the instruction is a rotation instruction, establishing a rotation coordinate system, and setting a VR head-mounted display as an original point of the rotation coordinate system; obtaining rotation data of the VR head-mounted display in the rotating coordinate system; and synchronizing the rotation data to the target model for execution. According to the invention, the eye movement cursor of the VR head-mounted display is used as the selection of the model, the model selection and the movement operation are convenient, the rotation of the model is controlled by the rotation of the VR head-mounted display equipment, and the control efficiency of the model is greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of virtual and reality technology, and in particular to a model control method, system and storage medium applied to VR head display interaction. Background Art

[0002] Virtual and reality technology is mainly based on computer technology, and uses and integrates the latest developments of various high technologies such as 3D graphics technology, multimedia technology, simulation technology, display technology, servo technology, etc., and uses computers and other equipment to create a virtual world with realistic 3D vision, touch, smell and other sensory experiences. Therefore, the display of resources is the most important technology in virtual and reality technology. Since users can move in the constructed virtual world when experiencing virtual and reality technology, it is necessary to ensure that resources can follow the user's pace during the rendering process to ensure the user's experience.

[0003] Generally, when experiencing a virtual scene, a VR headset is used for interactive operations within the scene. However, for some scenes that require custom model building, the model needs to be controlled. The traditional control method requires the use of a handle to select the model, and the movement or rotation of the model requires the use of the handle function. Not only does it add additional peripherals, but it is also difficult for users to operate and is not easy to get started. Therefore, it is not conducive to the development of virtual and reality technologies. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that when using a VR headset, a handle is needed to control the model, which increases the cost and difficulty of operation. A model control method, system and storage medium for VR headset interaction are proposed. The method, system and storage medium have the advantages of using the eye-moving cursor of the VR headset to select and quickly move the model, and using the gyroscope of the VR headset to rotate the model. The method and system are convenient and easy to control.

[0005] In a first aspect, the present invention achieves the above-mentioned purpose through the following technical solutions, a model control method applied to VR head display interaction, the method comprising the following steps: Monitoring the eye movement cursor of the VR head display, including monitoring the position coordinates and collision detection results of the eye movement cursor; After detecting that the eye movement cursor collides with the model, obtaining a user's selection instruction and selecting the model as a target model; Detecting an operation instruction of a user, wherein the operation instruction includes a movement instruction and a rotation instruction; If it is the movement instruction, acquiring the position coordinates of the eye cursor in real time, and moving the target model to the position coordinates after receiving a confirmation message; If it is the rotation instruction, then: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution.

[0006] Preferably, while acquiring the position coordinates of the eye-movement cursor in real time, a virtual model of the target model is generated in real time on the position coordinates.

[0007] Preferably, the method also includes setting function buttons on the VR headset, the function buttons including a selection button, a confirmation button and a function switching button, the selection button is used to send a selection instruction, the confirmation button is used to send a confirmation message, and the function switching button is used to switch between movement instructions and rotation instructions.

[0008] Preferably, the method further comprises setting an observation camera and binding an observation window on the field of view screen, and the method of binding the observation window on the field of view screen is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

[0009] Preferably, the method for obtaining the rotation data of the VR head display in the rotation coordinate system is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and angle of the corresponding axis each time the VR headset is rotated are recorded as rotation data.

[0010] Preferably, the method of synchronizing the rotation data to the target model for execution is: The current state of the target model is used as the origin of the rotation coordinate system, one direction of the target model is selected as the X-axis, and the direction perpendicular to the X-axis is selected as the Z-axis; Receive the rotation data, and rotate the target model according to the rotation direction and rotation angle in the rotation data; After receiving the confirmation message, the target model is fixed.

[0011] In a second aspect, the present invention achieves the above-mentioned purpose through the following technical solutions: a model control system applied to VR head display interaction, the system comprising: An eye-movement cursor monitoring unit, used to monitor the eye-movement cursor of the VR head display, including monitoring the position coordinates and collision detection results of the eye-movement cursor; A model selection unit, configured to obtain a user's selection instruction and select the model as a target model after detecting a collision between the eye cursor and the model; An operation monitoring unit, used to detect an operation instruction of a user, wherein the operation instruction includes a movement instruction and a rotation instruction; A movement execution unit, configured to obtain the position coordinates of the eye cursor in real time when the operation instruction is the movement instruction, and move the target model to the position coordinates after receiving a confirmation message; A rotation execution unit, used when the operation instruction is the rotation instruction: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution.

[0012] Preferably, the system further comprises a small window unit for setting an observation camera and binding the observation window on the field of view screen. The method for binding the observation window on the field of view screen is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

[0013] Preferably, the method for obtaining the rotation data of the VR head display in the rotation coordinate system in the rotation execution unit is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and angle of the corresponding axis each time the VR headset is rotated are recorded as rotation data.

[0014] In a third aspect, the present invention achieves the above-mentioned purpose through the following technical scheme: a storage medium having a computer program stored thereon, which, when the computer program is executed by a processor, implements the model control method applied to VR head display interaction as described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses the eye-movement cursor of the VR headset as a model selection method, so that the target selection follows the eye movement. Compared with the traditional handle selection, it has high accuracy and is more efficient to move. The eye-movement selection method can facilitate the selection of the model and the movement of the model, and then the rotation of the VR headset device is used to control the rotation of the model. There is no need to manually select the model for rotation, which greatly improves the control efficiency of the model.

[0016] 2. The present invention also generates an observation window, so that when the VR headset is rotated, the state of the model rotation can be observed in real time, thereby improving the accuracy of the rotation. There is no need to worry about the loss of the target model image when the VR headset is rotated, making the rotation control easy to use and simple to operate. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of the model control method applied to VR head display interaction of the present invention.

[0018] Figure 2 It is a schematic diagram showing the display of the observation window of the present invention on the visual field screen.

[0019] Figure 3 It is a schematic diagram showing the coordinate axes of the target model of the present invention and the coordinate axes of the VR head display.

[0020] Figure 4 This is a schematic diagram of the model control system used for VR head display interaction of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Example 1 like Figure 1 As shown, a model control method applied to VR head display interaction, the method comprises the following steps: Monitoring the eye cursor of the VR headset includes monitoring the position coordinates and collision detection results of the eye cursor. VR headsets generally have built-in eye cursors, which capture the user's visual focus through visual technology to perform subsequent interactive operations. In this step, the information of the eye cursor is obtained in order to obtain the position coordinates and collision detection results of the eye cursor. According to the position coordinates, the position of the user's visual focus in the scene can be grasped, and according to the collision detection results, the model that touches the user's visual focus can be obtained, thereby providing data support for subsequent interactions.

[0023] After detecting that the eye cursor collides with the model, the user's selection instruction is obtained, and the model is selected as the target model. This step uses the eye cursor to realize the selection function. The specific principle is to perform collision detection through the eye cursor. During the detection process, the eye cursor will use the user's gaze as a ray to contact the model in the scene until the user's selection instruction is received. The current ray collides with the model as the target model. This method is to select the model pre-placed in the scene. Another method of selecting the target model is to display the model in the form of an image list on the display interface, and use the eye cursor to select a specific model as the target model. Both methods can be selected according to the actual application scenario, and there are various ways to select the model, which are not limited to these two methods. All selection methods will not be described here one by one.

[0024] Detect the user's operation instructions, which include movement instructions and rotation instructions. In model control, they specifically include model movement and rotation. Therefore, the operation instructions provided in this step include movement instructions and rotation instructions. During operation, the instructions issued by the user are selected. The following describes in detail how different operation instructions realize model control.

[0025] In the first case, if it is the moving instruction, the position coordinates of the eye cursor are obtained in real time, and the target model is moved to the position coordinates after receiving the confirmation message. The moving instruction is to move the selected target model. Since the position coordinates of the eye cursor can be directly obtained, the accurate coordinates of the placement point of the target model can be directly obtained. At this time, after the user sends a confirmation message, the position coordinates can be synchronized to the target model, and the target model is generated at the position coordinates. In specific applications, problems such as the placement area does not match the model size, and there are obstacles and occlusions may be encountered. Therefore, the moving operation needs to be refined during implementation. The target model is aligned with the center point and the position coordinates. If the placement area does not match the model size, a placement failure prompt will be generated. If there is partial occlusion, the target model will be randomly offset for placement, and the point closest to the position coordinates will be selected for placement.

[0026] In the second case, if it is the rotation instruction, then: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution. The rotation operation is different from the movement operation. As can be seen from the above, the movement operation is to move the target model to the specified position by obtaining the position coordinates of the eye cursor. As for the rotation operation, since this solution adopts the operation method of separating from the handle, it is not easy to obtain the flip angle of the target model. Therefore, the VR headset is used as the rotation origin, and the VR headset is rotated as a reference for the rotation of the target model. The VR headset is generally equipped with components such as a gyroscope, so it is easy to obtain accurate rotation data, and then the rotation data is synchronized to the target model for execution, so that the target model can be rotated to the desired state. The above two control methods can be used to control the model in the scene.

[0027] When performing a moving operation, in order to more intuitively observe whether the moved position of the target model matches the target position, the position coordinates of the eye cursor are acquired in real time and a virtual model of the target model is generated in real time on the position coordinates. The virtual model allows the user to observe in advance whether the target model is suitable when placed at the position coordinates. By pre-displaying the virtual model, the problem of re-performing the moving operation due to inappropriate placement of the target model can be reduced, thereby greatly improving the efficiency of the moving operation.

[0028] The method also includes setting function buttons on the VR head display, the function buttons include a selection button, a confirmation button and a function switching button, the selection button is used to send a selection instruction, the confirmation button is used to send a confirmation message, and the function switching button is used to switch between movement instructions and rotation instructions. The function buttons can be integrated into the VR head display, and the user sends corresponding instructions by manually pressing or touching. For example, when the target model is selected, the selection button can help the user select the target model, and when performing a movement or rotation operation, the confirmation button can determine the movement position or rotation degree. The function switching button can be a switch button to realize the switching of movement and rotation instructions, or it can be two buttons, which respectively map corresponding movement instructions and rotation instructions. The above-mentioned function buttons can also be virtual buttons, and whether the physical button or the virtual button is selected according to the design of the VR head display.

[0029] like Figure 2 As shown, the method further includes setting an observation camera and binding an observation window on the field of view screen. The method of binding the observation window on the field of view screen is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

[0030] The observation window is displayed on the field of view when the operation instruction is a rotation instruction, and the observation window always moves with the movement of the picture, so that no matter how the user rotates, he can see the content played in the observation window, and the observation window plays the picture of the target model, so that the user can see the target model rotating with the rotation of the VR headset. The VR headset is generally worn on the head, so adding an observation window can make it convenient for the user to observe the target model following the rotation when the head is turned, without having to look for the target model in the scene, thereby improving the convenience of use.

[0031] As mentioned above, during the rotation operation, it is necessary to obtain the rotation data of the VR head display device. In order to reduce the error of the rotation data, the method for obtaining the rotation data of the VR head display in the rotation coordinate system is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and rotation angle of the corresponding axis are recorded each time the VR headset is rotated as rotation data. In this method, only the X-axis motion information and Z-axis motion information of the gyroscope are obtained. The X-axis motion information can obtain the horizontal rotation angle and direction of the VR headset, and the Z-axis motion information can obtain the vertical rotation angle and direction of the VR headset. The rotation of the target model is achieved through the four rotation directions of up, down, left, and right. Although the rotation direction can only be rotated in four directions, the target model requires complex rotation angles, and may need to be rotated multiple times, which is cumbersome to operate. However, the advantage of this design is that because the VR headset is worn on the head, the rotation angle of the neck cannot be too large or more complicated, so this design is more in line with the actual user's head rotation. Moreover, when rotating the VR headset, it is easy to have offsets in other directions, which cannot be accurately controlled. Here, only the rotation angles in four directions are obtained, which can filter out small offsets and make the rotation of the target model more accurate.

[0032] The above describes in detail how the rotation data is obtained. Here, how the target model executes the rotation data is further explained. The method of synchronizing the rotation data to the target model for execution is: The current state of the target model is used as the origin of the rotation coordinate system, one direction of the target model is selected as the X-axis, and the direction perpendicular to the X-axis is selected as the Z-axis; Receive the rotation data, and rotate the target model according to the rotation direction and rotation angle in the rotation data; After receiving the confirmation message, the target model is fixed. Each time the target model executes the rotation data, only the current state is used as the original position of the rotation coordinate system, for example Figure 3As shown, the target model at this time is not in the straightened state, and the X-axis and Z-axis are not the coordinate axes in the straightened state. The rotation data is when the VR headset is in the straightened state. It can be seen in the figure that the X-axis and Z-axis of the VR headset are in the straightened state. This is to ensure that when the user rotates the target model for the second time, there is no need to restore the VR headset to a state consistent with the current target model. When the VR headset is rotated, the recorded rotation data is executed by the target model, and the target model is executed according to the coordinate axes of the current state. In this way, no matter how many times the target model needs to be rotated, the VR headset can be rotated in the initial state, reducing the rotation angle of the neck and satisfying the actual experience of the user.

[0033] Example 2 like Figure 4 As shown, a model control system applied to VR head display interaction, the system includes: An eye-movement cursor monitoring unit, used to monitor the eye-movement cursor of the VR head display, including monitoring the position coordinates and collision detection results of the eye-movement cursor; A model selection unit, configured to obtain a user's selection instruction and select the model as a target model after detecting a collision between the eye cursor and the model; An operation monitoring unit, used to detect an operation instruction of a user, wherein the operation instruction includes a movement instruction and a rotation instruction; A movement execution unit is used to obtain the position coordinates of the eye cursor in real time when the operation instruction is the movement instruction, and move the target model to the position coordinates after receiving a confirmation message; and at the same time, generate a virtual model of the target model in real time at the position coordinates.

[0034] A rotation execution unit, used when the operation instruction is the rotation instruction: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution.

[0035] Function buttons are set on the VR head display, and the function buttons include a selection button, a confirmation button and a function switching button. The selection button is used to send a selection instruction, the confirmation button is used to send a confirmation message, and the function switching button is used to switch between a movement instruction and a rotation instruction.

[0036] The system also includes a small window unit, which is used to set an observation camera and bind the observation window on the field of view. The method of binding the observation window on the field of view is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

[0037] The method for obtaining the rotation data of the VR head display in the rotation coordinate system in the rotation execution unit is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and angle of the corresponding axis each time the VR headset is rotated are recorded as rotation data.

[0038] The method of synchronizing the rotation data to the target model for execution is: The current state of the target model is used as the origin of the rotation coordinate system, one direction of the target model is selected as the X-axis, and the direction perpendicular to the X-axis is selected as the Z-axis; Receive the rotation data, and rotate the target model according to the rotation direction and rotation angle in the rotation data; After receiving the confirmation message, the target model is fixed.

[0039] Embodiment 2 is essentially the same as Embodiment 1, so the specific working principles of each unit will not be described in detail.

[0040] Example 3 This embodiment proposes a storage medium, which includes a program storage area and a data storage area, wherein the program storage area can store an operating system and programs required for running instant messaging functions, etc.; the data storage area can store various instant messaging information and operation instruction sets, etc. The computer program is stored in the program storage area, and when the computer program is executed by the processor, the model control method for VR head display interaction as described in Example 1 is implemented. The processor may include one or more central processing units, or a digital processing unit, etc.

[0041] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

[0042] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A model control method for VR head display interaction, characterized in that: The method comprises the following steps: Monitoring the eye movement cursor of the VR head display, including monitoring the position coordinates and collision detection results of the eye movement cursor; After detecting that the eye movement cursor collides with the model, obtaining a user's selection instruction and selecting the model as a target model; Detecting an operation instruction of a user, wherein the operation instruction includes a movement instruction and a rotation instruction; If it is the moving instruction, acquiring the position coordinates of the eye cursor in real time, and moving the target model to the position coordinates after receiving a confirmation message; If it is the rotation instruction, then: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution.

2. The model control method for VR head display interaction according to claim 1, characterized in that: While acquiring the position coordinates of the eye-movement cursor in real time, a virtual model of the target model is generated in real time on the position coordinates.

3. The model control method for VR head display interaction according to claim 1, characterized in that: The method also includes setting function buttons on the VR head display, wherein the function buttons include a selection button, a confirmation button and a function switching button, the selection button is used to send a selection instruction, the confirmation button is used to send a confirmation message, and the function switching button is used to switch between a movement instruction and a rotation instruction.

4. The model control method for VR head display interaction according to claim 1, characterized in that: The method further includes setting an observation camera and binding an observation window on the field of view screen. The method of binding the observation window on the field of view screen is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

5. The model control method for VR head display interaction according to claim 1, characterized in that: The method for obtaining the rotation data of the VR head display in the rotation coordinate system is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and angle of the corresponding axis each time the VR headset is rotated are recorded as rotation data.

6. The model control method for VR head display interaction according to claim 5, characterized in that: The method of synchronizing the rotation data to the target model for execution is: The current state of the target model is used as the origin of the rotation coordinate system, one direction of the target model is selected as the X-axis, and the direction perpendicular to the X-axis is selected as the Z-axis; Receive the rotation data, and rotate the target model according to the rotation direction and rotation angle in the rotation data; After receiving the confirmation message, the target model is fixed.

7. A model control system applied to VR head display interaction, characterized in that: The system includes: An eye-movement cursor monitoring unit, used to monitor the eye-movement cursor of the VR head display, including monitoring the position coordinates and collision detection results of the eye-movement cursor; A model selection unit, configured to obtain a user's selection instruction and select the model as a target model after detecting a collision between the eye cursor and the model; An operation monitoring unit, used to detect an operation instruction of a user, wherein the operation instruction includes a movement instruction and a rotation instruction; A movement execution unit, configured to obtain the position coordinates of the eye cursor in real time when the operation instruction is the movement instruction, and move the target model to the position coordinates after receiving a confirmation message; A rotation execution unit, used when the operation instruction is the rotation instruction: Establishing a rotating coordinate system, and setting the VR head display as the origin of the rotating coordinate system; Obtaining rotation data of the VR head display in the rotation coordinate system; The rotation data is synchronized to the target model for execution.

8. The model control system for VR head display interaction according to claim 7, characterized in that: The system also includes a small window unit, which is used to set an observation camera and bind the observation window on the field of view. The method of binding the observation window on the field of view is: Set the observation window to always be fixed on the field of view; When a rotation instruction is detected, the observation window plays the image of the target model captured by the observation camera.

9. The model control system for VR head display interaction according to claim 7, characterized in that: The method for obtaining the rotation data of the VR head display in the rotation coordinate system in the rotation execution unit is: Obtaining gyroscope information of the VR head display, wherein the gyroscope information includes X-axis motion information and Z-axis motion information; The rotation direction and angle of the corresponding axis each time the VR headset is rotated are recorded as rotation data.

10. A storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the model control method applied to VR head display interaction as described in any one of claims 1 to 6 is implemented.

Citation Information

Patent Citations

  • On-site cognitive method and device for internal structure of electric power device

    CN106910244A

  • Eye movement tracking-based man-machine interactive system and working method thereof

    CN107656613A

  • Man-machine interaction system, man-machine interaction method, wearable device and head display device

    CN115480639A

  • Equipment maintenance method, device and system

    CN116468206A

  • Interface interaction method, device and equipment of AR (Augmented Reality) equipment and storage medium

    CN116991233A