A self-calibration method for positioning camera installation parameters in a virtual reality device

Abstract

The present application relates to a kind of self-calibration methods of positioning camera installation parameters in virtual reality equipment, utilize 3D display, 3D glasses or interactive pen in system, visual positioning component, without the aid of external calibration equipment, positioning camera in system installation parameter calibration can be realized.First, display mark point on display and record coordinate position, then, based on the principle of visual measurement, the coordinates of the mark point in the coordinate system of positioning camera are solved, according to the corresponding relationship of at least 3 groups of space points in the coordinate system of display and camera, the calibration of positioning camera installation parameter in system can be realized.The calibration method is convenient to build environment, without the assistance of professional personnel and professional equipment, the operation steps are simple, the calibration accuracy meets the application demand, enhances the practicability and maintainability of desktop virtual reality system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of virtual reality, and relates to a self-calibration method for positioning camera installation parameters in a virtual reality device. BACKGROUND

[0002] Virtual reality (VR) technology is an important direction of today's technological development, and a desktop-level virtual reality product is an important branch of the virtual reality field. The system mainly consists of a 3D display, a graphics workstation, 3D glasses, an interactive pen, a positioning assembly and the like. When the product is working, the system tracks the visual angle information of the user and the motion information of the user operating the interactive pen in real time through the positioning assembly, generates corresponding 3D graphics according to the visual angle and motion of the user, and displays the 3D graphics through the 3D display, so that a good human-computer interaction experience effect can be achieved. The desktop-level virtual reality system usually adopts a visual measurement method to track and position the user operation information. The positioning assembly is composed of a plurality of cameras which are distributed around the display. The 3D glasses and the interactive pen are provided with positioning marker points. The system controls the cameras to shoot the marker point images and processes the images. According to the computer vision theory, the position and posture information of the 3D glasses and the interactive pen can be solved, and then the user operation information can be obtained.

[0003] For the desktop-level virtual reality system, 3D image display is based on the display coordinate system, and the position and posture data captured by the positioning assembly is based on the visual system coordinate system. Therefore, in order to realize correct virtual interaction operation, it is necessary to first complete the calibration of the installation parameters of the positioning camera in the system, that is, the calibration of the position and posture relationship of the positioning camera relative to the display. For the calibration of the parameters, a special target body or an external camera is usually needed to establish a corresponding relationship between the two coordinate systems. However, these calibration methods need professional personnel and special equipment, and the environment is complex to build, the steps are cumbersome, and the comprehensive cost of calibration is high, which to some extent affects the popularization and application of the desktop-level virtual reality system. SUMMARY

[0004] Technical problems to be solved

[0005] In order to avoid the shortcomings of the prior art, the present application proposes a self-calibration method for positioning camera installation parameters in a virtual reality device. According to the characteristics of the desktop-level virtual reality system, a self-calibration method for positioning camera installation parameters is proposed. The user can complete the calibration of the positioning camera installation parameters by himself without the aid of external targets or other equipment. The operation is simple and fast, and the calibration accuracy meets the virtual interaction requirements.

[0006] Technical scheme

[0007] A self-calibration method for positioning camera installation parameters in a virtual reality device, characterized in that the steps are as follows:

[0008] Step 1: Establish a display coordinate system OD X D Y D Z D , the center of the display is taken as the coordinate center O D , the display plane is taken as X D Y D , and the coordinate perpendicular to the display plane is taken as Z D ; a visual system coordinate system O V X V Y V Z V is established, the center of the positioning camera is taken as the coordinate center O V , the plane of the lens of the positioning camera is taken as X D Y D , and the coordinate perpendicular to the lens plane is taken as Z V ;

[0009] Step 2: taking any point on the display plane as a mark point, the coordinates of the mark point in the display coordinate system are (x D0 , y D0 , z D0 );

[0010] Step 3: aligning the mark point on the display with the visual axis direction of the 3D glasses or the pen axis direction of the interactive pen;

[0011] Step 4: obtaining the parametric equation of the axis of the visual system coordinate system O V X V Y V Z V and solving the pose parameters by using the visual positioning component,

[0012] Step 5: changing the position of the 3D glasses or the interactive pen, repeating steps 3-4, solving and recording at least 10 groups of axis pose parameters;

[0013] Step 6: obtaining the coordinates of the intersection of all axes, i.e. the coordinates of the mark point in the positioning camera coordinate system, by using an optimization method;

[0014] Step 7: changing the display position of the mark point on the display for n times, repeating steps 2-6 for n times, obtaining the coordinates of multiple mark points in the display coordinate system and the positioning camera coordinate system, and completing the pose relationship matrix between the display coordinate system and the positioning camera coordinate system, i.e. the installation parameters of the positioning camera, according to the theory of stereographic analytic geometry.

[0015] The parametric equation of the axis of the visual system coordinate system O V X V Y V Z V of step 4 is: i = 1, 2...m, p i、 q i , r i are coefficients of parametric equation.

[0016] The step 6 calculates the coordinates (x V0 , y V0 , z V0 ) of the marker points in the positioning camera coordinate system: establish the distance from the intersection point to all axes and establish the objective function: wherein Optimize (x V0 , y V0 , z V0 ) by using L-M or other optimization algorithms.

[0017] The step 7 calculates the installation parameters of the positioning camera: the pose relationship matrix between the display coordinate system and the positioning camera coordinate system is [R VD |T VD ], wherein R VD is a 3x3 matrix, which is the rotation relationship between the two coordinate systems, and T VD is a 3x1 vector, which is the translation relationship between the two coordinate systems; the coordinate values (x Di , y Di , z Di ) of the plurality of marker points in the display coordinate system and the corresponding coordinate values (x Vi , y Vi , z Vi ) in the positioning camera coordinate system are substituted into [x Vi , y Vi , z Vi ] T = R VD [x Di , y Di , z Di ] T + T VD , and the pose relationship matrix between the positioning camera and the display coordinate system, i.e. the installation parameters of the positioning camera, are solved by using the optimization algorithm.

[0018] The n is greater than or equal to 3.

[0019] The m is greater than 10.

[0020] Beneficial effects

[0021] The application provides a self-calibration method for positioning camera installation parameters in a virtual reality device, which utilizes a 3D display, 3D glasses or an interactive pen, a visual positioning component in the system, and does not need to use external calibration equipment to realize calibration of the positioning camera installation parameters in the system.

[0022] The application fully combines the characteristics of a desktop virtual reality system, and uses the components of the system to quickly and conveniently calibrate the positioning camera installation parameters, thereby enhancing the practicability and maintainability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 Fig. 1 is a schematic diagram of a typical desktop virtual reality system.

[0024] Figure 2 Fig. 2 is a schematic diagram of a display displaying a mark point.

[0025] Figure 3 Fig. 3 is a schematic diagram of a parameter self-calibration based on 3D glasses.

[0026] Figure 4 Fig. 4 is a schematic diagram of a parameter self-calibration based on a touch pen.

[0027] Fig. 1 is a schematic diagram of a typical desktop virtual reality system. DETAILED DESCRIPTION

[0028] The application will be further described in combination with embodiments and drawings:

[0029] The virtual reality device positioning camera installation parameter self-calibration method provided by the application mainly uses display equipment, 3D glasses or an interactive pen in the system, and does not need to use external equipment to complete calibration of the positioning camera installation parameters. Mark points are displayed on the display, a user operates the 3D glasses or the interactive pen, a corresponding relationship between coordinates of a plurality of space points in a display coordinate system and a positioning camera coordinate system is established based on visual measurement theory, and then calibration and solving of the position and posture parameters between the display coordinate system and the positioning camera coordinate system are completed.

[0030] Combination Figure 1A typical desktop virtual reality system is shown, which includes a 3D display, 3D glasses, an interactive pen, and a visual positioning assembly composed of four cameras, wherein the 3D glasses and the interactive pen are provided with positioning mark points. The following describes the self-calibration method in detail. Figure 3 And Figure 4 The self-calibration method is described in detail.

[0031] The calibration steps are as follows:

[0032] 1. Display an obvious mark point on the display, such as Figure 2 Record the coordinates (x D , y D , z D ) of the point in the display coordinate system O D -X D0 Y D0 Z D0 ;

[0033] 2. The operator wears 3D glasses and changes the head posture to align the visual axis direction with the mark point on the display, as shown in Figure 3 Or the operator manipulates the interactive pen to align the pen axis direction with the mark point on the display, as shown in Figure 4 ;

[0034] 3. The visual positioning assembly is used to solve the pose of the visual axis of the glasses or the pen axis, and the parameter equation of the axis in the visual system coordinate system O V -X V Y V Z V is obtained;

[0035] 4. The operator moves the 3D glasses or the stylus position, repeats steps 2-3, and records multiple sets of parameter equations of different pose axes i=1,2…n, x Vi , y Vi , z Vi , p i , q i , r i are the parameters of the space straight line equation;

[0036] 5. Solve the intersection coordinates (x V0 , y V0 , z V0 ) of the above axes, that is, the coordinate values of the display mark point in the positioning camera coordinate system. Theoretically, the intersection (x V0 , y V0 , z V0 ) satisfies But due to measurement errors, the target function is established by the distance from the intersection to all axes:

[0037] wherein

[0038] Optimization algorithm such as L-M or other optimization algorithm is used to solve (x V0 ,y V0 ,z V0 ), the more the axis, the more accurate the result is;

[0039] 6. Change the display position of the mark points on the display, repeat the above steps 1-5, record the coordinate values (x Di ,y Di ,z Di ) of at least three groups of mark points under the display coordinate system and the corresponding coordinate values (x Vi ,y Vi ,z Vi ) under the positioning camera coordinate system; set the pose relationship matrix between the display coordinate system and the positioning camera coordinate system as [R VD |T VD ], wherein R VD is a 3*3 matrix, describing the rotation relationship between the two coordinate systems, and T VD is a 3*1 vector, describing the translation relationship between the two coordinate systems; the above coordinate values are substituted into [x Vi ,y Vi ,z Vi ] T =R VD [x Di ,y Di ,z Di ] T +T VD , and the pose relationship matrix between the positioning camera and the display coordinate system, i.e. the installation parameters of the positioning camera, can be solved by using an optimization algorithm, and the more the coordinate pairs, the more accurate the calibration result is.

[0040] Up to now, the system parameter self-calibration is completed.

[0041] The above merely illustrates one specific embodiment of the present application, and is not used to limit the protection scope of the present application, and any modification, equivalent replacement and improvement within the idea and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A self-calibration method for positioning camera installation parameters in a virtual reality device, characterized in that... The steps are as follows: Step 1: Establish the monitor coordinate system With the center of the monitor as the coordinate center The display plane is The coordinates perpendicular to the display plane are Establish a coordinate system for the vision system. The center of the positioning camera is used as the coordinate center. Positioning the camera lens plane as The coordinates perpendicular to the lens plane are ; Step 2: Take any point on the screen plane as a marker point. The coordinates of this point in the screen coordinate system are: ; Step 3: Align the 3D glasses' visual axis or the pen's axis with the marker on the display. Step 4: Obtain the visual system coordinate system by using the visual positioning component to align the eye's visual axis or the interactive pen's axis. The parametric equations of the lower axis are derived, and the pose parameters are solved; the vision system coordinate system is defined. The parametric equation of the lower axis is: , , These are the coefficients of the parametric equations; Step 5: Change the position of the 3D glasses or interactive pen, repeat steps 3-4, and solve and record at least 10 sets of axis pose parameters; Step 6: Use optimization methods to obtain the coordinates of the intersection points of all axes, i.e., the coordinates of the marker points in the positioning camera coordinate system; Calculation: Establish the distances from the intersection points to all axes and establish the objective function: ,in Using LM or other optimization algorithms to Perform optimization; Step 7: Change the display position of the markers on the monitor n times, and repeat steps 2 to 6 n times to obtain the coordinates of multiple markers in the monitor coordinate system and the positioning camera coordinate system. Based on solid analytical geometry theory, complete the pose relationship matrix between the monitor coordinate system and the positioning camera coordinate system, that is, the installation parameters of the positioning camera.

2. The self-calibration method for positioning camera installation parameters in a virtual reality device according to claim 1, characterized in that: Step 7 yields the calculation of the positioning camera's installation parameters: the pose relationship matrix between the display coordinate system and the positioning camera coordinate system is as follows. ,in, It is a 3×3 matrix, representing the rotation relationship between the two coordinate systems. This is a 3×1 vector representing the translation relationship between two coordinate systems; it represents the coordinates of multiple marker points in the display coordinate system. and their corresponding coordinates in the positioning camera coordinate system. Substitution The optimization algorithm is used to solve for the pose relationship matrix between the positioning camera and the display coordinate system, which is the installation parameters of the positioning camera.

3. The self-calibration method for positioning camera installation parameters in a virtual reality device according to claim 1, characterized in that: The nth time .

4. The self-calibration method for positioning camera installation parameters in a virtual reality device according to claim 1, characterized in that: The value of m is greater than 10.

Images