CN104102357A - Method and device for checking 3D (three-dimensional) models in virtual scenes - Google Patents

Abstract

The present invention is applicable to the field of model detection of human-computer interaction, and provides a 3D model detection method and device in a virtual scene. The method includes: corresponding to the spatial attitude angle of the interactive pen, and mapping it in real time in the virtual scene Draw a virtual control line; corresponding to each 3D model in the virtual scene, construct a surrounding volume respectively, and the surrounding volume wraps the corresponding 3D model; judge the encirclement between the real-time mapped virtual control line and the 3D model Whether the body intersects; when the virtual control line intersects with the bounding volume of the 3D model, judge whether any triangular mesh in the 3D model corresponding to the virtual control line and the intersecting bounding volume intersects; When intersecting with any triangle mesh in the 3D model corresponding to the intersecting bounding volume, it is determined that the virtual control line intersects with the 3D model corresponding to the intersecting bounding volume, so as to determine that the 3D model is detected. The embodiment of the present invention can improve the detection accuracy of the 3D model.

Description

A 3D model detection method and device in a virtual scene

technical field

The invention belongs to the field of model detection of human-computer interaction, and in particular relates to a 3D model detection method and device in a virtual scene.

Background technique

With the in-depth research and application of virtual reality technology, the further development of 3D visualization display technology and human-computer interaction technology, people can render and design various virtual reality scenes, and carry out necessary control over the 3D models in the virtual scene. Before controlling the 3D model in the virtual scene, it is necessary to detect the 3D model in the virtual scene.

At present, the existing 3D model detection method in the virtual scene is mainly realized by the following method: establish a ray cluster, and then detect the intersection between the established ray cluster and the detection object, and when a certain ray is detected to intersect the detection object, Find the hit 3D model according to the mapping relationship between the object and the model. Since the model is only detected by detecting the intersection of the established ray cluster and the object, the detection accuracy is too low.

Contents of the invention

An embodiment of the present invention provides a method for detecting a 3D model in a virtual scene, aiming at solving the problem of low accuracy in detecting a 3D model in existing methods.

The embodiment of the present invention is achieved in this way, a 3D model detection method in a virtual scene, the method includes the following steps:

Corresponding to the spatial attitude angle of the interactive pen, a virtual control line is mapped in real time in the virtual scene;

Corresponding to each 3D model in the virtual scene, a surrounding volume is respectively constructed, the surrounding volume wraps the corresponding 3D model, and the 3D model is composed of a plurality of triangular meshes;

Judging whether the virtual control line of the real-time mapping intersects with the bounding volume of the 3D model;

When the virtual control line intersects with the bounding volume of the 3D model, it is judged whether the virtual control line intersects with any triangular mesh in the 3D model corresponding to the intersecting bounding volume;

When the virtual control line intersects any triangle mesh in the 3D model corresponding to the intersecting bounding volume, it is determined that the virtual control line intersects the 3D model corresponding to the intersecting bounding volume, so as to determine that the 3D model is detected.

Another object of the embodiments of the present invention is to provide a 3D model detection device in a virtual scene, the device comprising:

A virtual control line generating unit, configured to map a virtual control line in the virtual scene in real time corresponding to the spatial attitude angle of the interactive pen;

A model enclosure construction unit, configured to construct an enclosure corresponding to each 3D model in the virtual scene, the enclosure wraps the corresponding 3D model, and the 3D model is composed of a plurality of triangular meshes;

The intersection detection unit of the model bounding volume is used for judging whether the virtual control line of the real-time mapping intersects with the bounding volume of the 3D model;

The intersection detection unit of the model triangular mesh is used to determine whether any triangular mesh in the 3D model corresponding to the virtual control line and the intersecting bounding volume is intersect;

A model detection and determination unit, configured to determine that the virtual control line intersects the 3D model corresponding to the intersecting volume when the virtual control line intersects with any triangle mesh in the 3D model corresponding to the intersecting volume, so as to It is determined that a 3D model is detected.

In the embodiment of the present invention, since the bounding volume of the 3D model is detected first, after it is judged that the corresponding direction vector in the virtual scene intersects with the bounding volume of the 3D model after the interactive pen rotates, it is detected that the interactive pen rotates in the virtual scene. Whether the corresponding direction vector in the scene intersects with any triangular mesh in the 3D model, thus reducing the amount of computation, thereby effectively improving the speed and accuracy of detecting the 3D model in the virtual environment scene.

Description of drawings

Fig. 1 is a flowchart of a 3D model detection method in a virtual scene provided by the first embodiment of the present invention;

Fig. 2 is a schematic diagram of the attitude angle provided by the first embodiment of the present invention;

3 is a structural diagram of a 3D model detection device in a virtual scene provided by a second embodiment of the present invention;

Fig. 4 is a structural diagram of another 3D model detection device in a virtual scene provided by the second embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

In the embodiment of the present invention, according to the spatial attitude angle of the interactive pen, a virtual control line is mapped in real time in the virtual scene, and the bounding volume of the 3D model is determined according to the stored vertex data information of the 3D model, and then the virtual control line is judged Whether it intersects with the bounding volume of the 3D model, and when the virtual control line intersects with the bounding volume of the 3D model, it is judged whether the virtual control line intersects with any triangle mesh in the 3D model corresponding to the intersecting bounding volume, and if it intersects , to determine that a 3D model has been detected.

In order to illustrate the technical solutions of the present invention, specific examples are used below to illustrate.

Embodiment one:

Fig. 1 shows a flow chart of a 3D model detection method in a virtual scene provided by the first embodiment of the present invention, which is described in detail as follows:

Step S11 , mapping a virtual control line in real time in the virtual scene corresponding to the spatial attitude angle of the interactive pen.

In this step, the interactive pen refers to a pen-like operating tool used by the user to interact with the smart terminal, and the user can interact with the 3D model in the smart terminal by changing the posture of the interactive pen.

Wherein, the attitude angle of the interactive pen includes a pitch angle, a roll angle and a yaw angle of the interactive pen. See Figure 2. In Figure 2, the coordinate system composed of Xg axis, Yg axis, and Zg axis is the ground coordinate system. The Zg axis is perpendicular to the ground and points to the center of the earth. The Yg axis is perpendicular to the Xg axis in the horizontal plane, and its direction is right-handed. The rules are determined; the coordinate system composed of Xb axis, Yb axis and Zb axis is the coordinate system of the interactive pen, the Xb axis is in the plane of the interactive pen and parallel to the design axis of the interactive pen, and the Yb axis is perpendicular to the symmetry plane of the interactive pen and points to the right of the interactive pen The Zb axis is in the symmetry plane of the interactive pen, perpendicular to the Xb axis and pointing downward of the interactive pen. The pitch angle refers to the angle between the Xb axis of the interactive pen coordinate system and the horizontal plane. When the positive semi-axis of the Xb axis is above the horizontal plane passing through the coordinate origin, the pitch angle is positive, otherwise it is negative; the yaw angle refers to the interactive pen The angle between the projection of the Xb axis of the coordinate system on the horizontal plane and the Xg axis of the ground coordinate system, when the Xg axis is turned counterclockwise to the projection of the Xb axis, the yaw angle is positive, otherwise it is negative; the roll angle is The angle between the Zb axis of the interactive pen coordinate system and the vertical plane passing through the Xb axis is positive when the interactive pen rolls to the right, and negative when the interactive pen is rolled to the right.

Preferably, in step S11, corresponding to the spatial attitude angle of the interactive pen, before mapping a virtual control line in the virtual scene in real time, the following steps are included:

A1. Determine the initial pitch and roll angles. In this step, the initial pitch angle and roll angle are determined by the three-axis gyroscope and the three-axis accelerometer installed on the interactive pen.

A2. Iteratively determine the final pitch angle and roll angle according to the Kalman filter and the initial pitch angle and roll angle. Since the gyroscope has high precision, it will drift for a long time, while the accelerometer has poor dynamic precision, but there is no long-term drift. Therefore, in order to make full use of the advantages of the two, data fusion can be performed through Kalman filtering to obtain stable and accurate pitch and roll angles. In this step: the initial pitch angle and roll angle are represented by quaternion and used as the state quantity of the Kalman filter system. The attitude kinematics equation is used as the state transition equation of the Kalman filter, and the acceleration information is used as the observation information of the filter, and then the calculation method of the Kalman filter is used for iterative calculation to update the initial pitch angle and roll angle to an accurate and stable pitch angle and roll angle.

Among them, the quaternion is composed of real numbers plus three elements i, j, k, and they have the following relationship: i^2=j^2=k^2=ijk=-1.

A3. Determine the magnetic field strength component in the horizontal plane according to the final pitch angle and roll angle and the magnetic field strength components of the x-axis, y-axis, and z-axis of the coordinate system where the interactive pen is located. In this step, the magnetic field strength components of the x-axis, y-axis, and z-axis of the coordinate system where the interactive pen is located can be obtained through a three-axis magnetometer installed on the interactive pen. Assuming that the magnetic field strength components of the x-axis, y-axis, and z-axis measured by the three-axis magnetometer are h _x , h _y , h _z respectively, the final pitch angle is θ, and the roll angle is γ. Since the interactive pen The coordinate system xyz where it is located rotates around the x-axis, and then rotates around the y-axis. The xoy plane of the coordinate system where the interactive pen is located facilitates the coincidence of the horizontal plane XOY of the ground coordinate system. Therefore, using the above relationship, the following equation can be obtained:

$\left[\begin{array}{c}{h}_x\\ h_Y\\ h_Z\end{array}\right]=R_{\mathrm{pitch}}\&Center\; Dot;R_{\mathrm{roll}}\left[\begin{array}{c}{h}_x\\ h_{\mathrm{the\; y}}\\ h_z\end{array}\right],$ Among them, H _X , H _Y , H _Z are the magnetic field intensity components in the horizontal plane of the ground coordinate system, $R_{\mathrm{pitch}}=\left[\begin{array}{ccc}\cos \mathrm{\θ}& 0& -\sin \\ 0& 1& 0\\ \sin \mathrm{\θ}& 0& \cos \mathrm{\θ}\end{array}\right],R_{\mathrm{roll}}=\left[\begin{array}{ccc}1& 0& 0\\ 0& \cos \mathrm{\γ}& \sin \mathrm{\γ}\\ 0& \sin \mathrm{\γ}& \cos \mathrm{\γ}\end{array}\right]$

R _pitch is the rotation matrix obtained by rotating the coordinate system xyz where the interactive pen is located by an angle θ around the y-axis, and R _roll is the rotation matrix obtained by rotating the coordinate system xyz where the interactive pen is located by an angle γ around the x-axis. Expanding the above equation, we get:

H _X ＝h _x ·cosγ+h _y ·sinγ·sinθ+h _z ·cosγ·sinθ

H _Y ＝h _y ·cosγ-h _z ·sinγ

H _Z ＝-h _x sinθ+h _y sinγcosθ+h _z cosθcosγ

The "·" above refers to the inner product, and H _X and H _Y are obtained through the above three formulas.

A4. Determine the yaw angle according to the determined magnetic field intensity component in the horizontal plane.

Yaw angle=arctan(H _Y /H _X ), where H _X and H _Y are the magnetic field intensity components in the horizontal plane of the ground coordinate system.

After the spatial attitude angle of the interactive pen is determined, a virtual control line is mapped in real time in the virtual scene corresponding to the spatial attitude angle of the interactive pen, which specifically includes the following steps:

B1. Determine the rotation matrix of the interactive pen according to the attitude angle of the interactive pen. Specifically: B11. Determine the matrix obtained after the coordinate system where the interactive pen is located is rotated around the y-axis by an angle θ, where θ is the pitch angle. It can be seen from step S11 that the matrix obtained after the coordinate system where the interactive pen is located is rotated by an angle θ around the y-axis is $R_{\mathrm{pitch}}=\left[\begin{array}{ccc}\cos \mathrm{\θ}& 0& -\sin \\ 0& 1& 0\\ \sin \mathrm{\θ}& 0& \cos \mathrm{\θ}\end{array}\right].$ B12. Determine the matrix obtained after the coordinate system where the interactive pen is located is rotated around the z-axis by an angle α, where α is a yaw angle. Assuming that the coordinate system where the interactive pen is located is rotated by an angle α around the z-axis, the matrix obtained is R _z (α), then $R_z\left(\mathrm{\α}\right)=\left[\begin{array}{ccc}\cos \mathrm{\α}& -\sin & 0\\ \sin \mathrm{\α}& \cos \mathrm{\α}& 0\\ 0& 0& 1\end{array}\right].$ B13. Determine the matrix obtained after the coordinate system where the interactive pen is located is rotated around the x-axis by an angle γ, where γ is a roll angle. From step S11, it can be seen that the matrix obtained after the coordinate system where the interactive pen is located is rotated around the x-axis by an angle γ is $R_{\mathrm{roll}}=\left[\begin{array}{ccc}1& 0& 0\\ 0& \cos \mathrm{\γ}& \sin \mathrm{\γ}\\ 0& \sin \mathrm{\γ}& \cos \mathrm{\γ}\end{array}\right].$ B14. According to the determined matrix obtained by rotating the coordinate system of the interactive pen by an angle θ around the y-axis, the matrix obtained by rotating the coordinate system of the interactive pen by an angle α around the z-axis, and the coordinate system of the interactive pen around the x-axis The matrix obtained by rotating the axis by an angle of γ determines the rotation matrix of the interactive pen. Suppose the rotation matrix is R, then $\begin{array}{c}R=R_{\mathrm{roll}}\·R_z\left(\mathrm{\α}\right)\&Center\; Dot;R_{\mathrm{pitch}}=R_{\mathrm{roll}}\\ =\left[\begin{array}{ccc}1& 0& 0\\ 0& \cos \mathrm{\γ}& \sin \mathrm{\γ}\\ 0& \sin \mathrm{\γ}& \cos \mathrm{\γ}\end{array}\right]\left[\begin{array}{ccc}\cos \mathrm{\α}& -\sin \mathrm{\α}& 0\\ \sin \mathrm{\α}& \cos \mathrm{\α}& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{ccc}\cos \mathrm{\θ}& 0& -\sin \mathrm{\θ}\\ 0& 1& 0\\ \sin \mathrm{\θ}& 0& \cos \mathrm{\θ}\end{array}\right]\end{array}$ Expand to get: $R=\left[\begin{array}{ccc}\cos \mathrm{\θ}\·\cos \mathrm{\α}& \cos \mathrm{\θ}\·(-\sin \mathrm{\α})\·\cos \mathrm{\γ}-\sin \mathrm{\θ}\&Center\; Dot;\sin \mathrm{\γ}& \cos \mathrm{\θ}\&Center\; Dot;\sin \mathrm{\α}\&Center\; Dot;\sin \mathrm{\γ}-\sin \mathrm{\θ}\·\cos \mathrm{\γ}\\ \sin \mathrm{\α}& \cos \mathrm{\α}\·\cos \mathrm{\γ}& \cos \mathrm{\α}\&Center\; Dot;(-\sin \mathrm{\γ})\\ \sin \mathrm{\θ}\&Center\; Dot;\cos \mathrm{\α}& \sin \mathrm{\θ}\&Center\; Dot;(-\sin \mathrm{\α})\·\cos \mathrm{\γ}+\cos \mathrm{\θ}\&Center\; Dot;\sin \mathrm{\γ}& \sin \mathrm{\θ}\&Center\; Dot;\sin \mathrm{\α}\&Center\; Dot;\sin \mathrm{\γ}+\cos \mathrm{\θ}\·\cos \mathrm{\γ}\end{array}\right].$

B2. According to the rotation matrix of the interactive pen, the preset origin in the virtual scene, and the preset initial direction vector of the interactive pen, render the virtual control line corresponding to the rotation of the interactive pen in the virtual scene in real time. In this step, it is assumed that the initial direction vector of the preset interactive pen is After the interactive pen is rotated, the corresponding direction vector in the virtual scene (that is, the real-time mapped virtual control line) is R is a rotation matrix, then Where "·" represents the inner product operation. If the origin in the virtual scene is determined, the position of the corresponding direction vector (that is, the virtual control line mapped in real time) in the virtual scene after the interactive pen is rotated can be determined. Specifically, at the preset origin in the virtual scene, the corresponding direction vector in the virtual scene after the interactive pen is rotated is drawn as a virtual control line of a specified length. In this step, the graphics rendering tool OpenGL can be used to rotate the origin of the preset virtual scene along the interactive pen to the corresponding direction vector in the virtual scene Direction, draw a virtual control line of a certain length, so that the spatial attitude information of the interactive pen can be presented in real time through the virtual control line, so that the user can intuitively know the current state of the interactive pen.

Step S12, corresponding to each 3D model in the virtual scene, construct a bounding volume respectively, the bounding volume encloses the corresponding 3D model, and the 3D model is composed of a plurality of triangular meshes.

In this step, since each 3D model is composed of multiple triangular meshes, the vertex data information of the 3D model is the vertex data information corresponding to the multiple triangular meshes forming the 3D model. When constructing a bounding volume of a 3D model, compare the data information of each fixed point in the 3D model to determine which coordinate is the maximum vertex coordinate and the minimum vertex coordinate of the 3D model on the x-axis; determine which coordinate is the maximum vertex coordinate of the 3D model on the y-axis , Minimum vertex coordinates; determine which coordinates are the maximum vertex coordinates and minimum vertex coordinates of the 3D model on the z-axis. After the 6 coordinates listed above are determined, the 6 coordinates are used as the vertex coordinates of the bounding volume of the 3D model.

Step S13, judging whether the real-time mapped virtual control line intersects with the bounding volume of the 3D model.

Wherein, the step of judging whether the real-time mapped virtual control line intersects with the bounding volume of the 3D model specifically includes:

C1. Determine the function f(x, y, z) corresponding to the virtual control line, and the component of any point of f(x, y, z) on the x-axis, y-axis, and z-axis is f _x (x, y, z ), f _y (x,y,z), f _z (x,y,z). In this step, the function f(x, y, z) corresponding to the virtual control line passes through the preset origin in the virtual scene.

C2. Determine the components f x (x, y, z), f _y (x, y, z) and f z ( _x , y) of any point in f (x, y, z) on the x-axis, y-axis, and _z -axis , z) all meet the following conditions:

min x ≤ f _x (x, y, z) ≤ max x;

min y≤f _y (x,y,z)≤max y;

min z≤f _z (x,y,z)≤max z;

Wherein, the min x, max x, min y, max y, min z, max z are the vertex coordinates of the bounding volume. In this step, min x, max x, min y, max y, min z, max z are respectively the minimum vertex coordinates and maximum vertex coordinates of the 3D model on the x-axis; the minimum vertex coordinates and the maximum vertex coordinates on the y-axis; The minimum and maximum vertex coordinates of the z-axis.

C3 _. Components f x (x, y, z), f _y (x, y, z), f _z (x, y ,z) When the above conditions are all satisfied, it is determined that the virtual control line intersects the bounding volume of the 3D model; otherwise, it is determined that the virtual control line does not intersect the bounding volume of the 3D model. In this step, once it is detected that the virtual control line intersects with the bounding volume of a certain 3D model, the possibility that the virtual control line intersects with other remaining 3D models can be ruled out, that is, it is not necessary to detect the boundary between the virtual control line and other 3D models Whether the volume intersects, thereby greatly speeding up the detection speed of the 3D model, and because the intersection of the virtual control line and the bounding volume is judged according to the coordinates of the vertices of the virtual control line and the coordinates of the bounding volume of the 3D model, the judgment result will not Affected by whether the 3D model is occluded, the detection efficiency of the occluded 3D model in the virtual scene can be improved.

Step S14, when the virtual control line intersects with the bounding volume of the 3D model, it is judged whether the virtual control line intersects with any triangular mesh in the 3D model corresponding to the intersected bounding volume.

After it is determined to intersect with the bounding volume of a certain 3D model, the specific intersection detection between the virtual control line and the triangular mesh of the 3D model is performed next.

Wherein, when the virtual control line intersects with the bounding volume of the 3D model, the step of judging whether the virtual control line intersects with any triangular mesh in the 3D model corresponding to the intersecting bounding volume specifically includes:

Assume that the parametric equation of the triangle corresponding to any triangle mesh in the 3D model is (1-uv)V ₀ +uV ₁ +vV ₂ , where (1-uv), u, and v are the weights of the three vertices of the triangle; V ₀ , V ₁ , V ₂ are the coordinates of the three vertices of the triangle;

According to the origin in the preset virtual scene, the virtual control line and the three vertex coordinates of any triangular mesh in the 3D model, determine whether u and v meet the following conditions: u≥0, v≥0, u+ v≤1, if satisfied, determine that the virtual control line intersects with any triangle mesh in the 3D model corresponding to the intersecting bounding volume, otherwise, determine that the virtual control line intersects with the 3D model corresponding to the intersecting bounding volume Any of the triangular meshes in are disjoint.

Assuming that the origin in the preset virtual scene is O, and the direction of the virtual control line is D, then the virtual control line can be expressed as O+D·t (this expression can be compared with the f(x,y) appearing in the above C1 represents the control line), t≠0. Since the vertices of each triangular mesh of the 3D model are stored in the vertex array, each triangular mesh can be sequentially checked for intersection according to the storage order of the vertices. Suppose the vertices of a triangular mesh are V ₀ , V ₁ , V ₂ respectively, and u, v are the weights of the vertices V ₁ , V ₂ respectively, then the parametric equation of the triangular mesh can be expressed as (1-uv)V ₀ +uV ₁ +vV ₂ , when u≥0, v≥0, u+v≤1, it is determined to intersect with the triangular mesh. To judge whether u and v meet the above conditions, you need to determine the values of u and v first, and specifically determine the values of u and v through the following steps:

Calculate the unknown number t according to the equation O+D·t=(1-uv)V ₀ +uV ₁ +vV ₂ , if t is not zero, it means that the virtual control line intersects with the triangular mesh, that is, the specific 3D model up. According to O+D·t=(1-uv)V ₀ +uV ₁ +vV ₂ , the linear equations can be obtained:

$\left[\begin{array}{ccc}-\mathrm{D.}& V_1-V_0& V_2-V_0\end{array}\right]\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=o-V_0,$ Let E ₁ =V ₁ -V ₀ , E ₂ =V ₂ -V ₀ , T=oV ₀ to get $\left[\begin{array}{ccc}-\mathrm{D.}& {\mathrm{E.}}_1& {\mathrm{E.}}_2\end{array}\right]\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=T,$ by Cramer's law $\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=\frac{1}{\begin{array}{ccc}|-\mathrm{D.}& {\mathrm{E.}}_1& {\mathrm{E.}}_2\end{array}|}\left[\begin{array}{ccc}T& {\mathrm{E.}}_1& {\mathrm{E.}}_2\\ -\mathrm{D.}& T& {\mathrm{E.}}_2\\ -\mathrm{D.}& {\mathrm{E.}}_1& T\end{array}\right]$

Among them, "||" represents the determinant of the matrix.

Then according to the mixed product formula |a b c|=a×b·c=-a×c·b, where “×” means outer product calculation, and “·” means inner product calculation. $\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=\frac{1}{\begin{array}{ccc}|-\mathrm{D.}& {\mathrm{E.}}_1& {\mathrm{E.}}_2\end{array}|}\left[\begin{array}{ccc}T& {\mathrm{E.}}_1& {\mathrm{E.}}_2\\ -\mathrm{D.}& T& {\mathrm{E.}}_2\\ -\mathrm{D.}& {\mathrm{E.}}_1& T\end{array}\right]$ The formula can be changed to:

$\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=\frac{1}{|\mathrm{D.}\×{\mathrm{E.}}_2\&Center\; Dot;{\mathrm{E.}}_1|}\left[\begin{array}{c}T\×{\mathrm{E.}}_1\&Center\; Dot;{\mathrm{E.}}_2\\ \mathrm{D.}\×{\mathrm{E.}}_2\&Center\; Dot;T\\ T\×{\mathrm{E.}}_1\&Center\; Dot;\mathrm{D.}\end{array}\right],$ Let P=D×E ₂ , Q=T×E ₁ ,

can eventually be reduced to $\left[\begin{array}{c}t\\ u\\ v\end{array}\right]=\frac{1}{\left|P{\&Center\; Dot;\mathrm{E.}}_1\right|}\left[\begin{array}{c}Q\·{\mathrm{E.}}_2\\ P\&Center\; Dot;T\\ Q\&Center\; Dot;\mathrm{D.}\end{array}\right]$

According to the final simplified formula above, the calculation formulas of u and v can be obtained, and then the following method can be used to confirm whether u and v meet the condition of "u≥0, v≥0, u+v≤1":

Let ε be a positive floating point number very close to 0, let det=|P·E ₁ |, if det<ε, immediately end the intersection judgment with the current triangular mesh, and enter the judgment with the next triangular mesh. If det≥ε, set u′=P·T, if u′<0 or u′>det, it is judged that the finally calculated u value does not meet the conditions, then immediately jump out of the current judgment and enter the next triangular mesh judgment. If u'≥0 or u'≤det, set v'=Q·D, if v'<0 or u'+v'>det, it is determined that the final calculated value of v does not meet the conditions, and immediately jump out of the current judgment , enter the judgment with the next triangular mesh. Otherwise, if det≥ε, u'≥0 or u'≤det, and v'≤0 or u'+v'≤det, it can be judged that the virtual control line intersects with a specific triangular mesh.

Since the same triangular mesh has front and back sides, during the detection process with the triangular mesh, only the intersection with the front of the triangular mesh is detected, which can save half the detection time. When an intersection with a specific triangular mesh is detected, Record the intersection point and the number of the bounding volume corresponding to the 3D model, and the detection is completed.

Step S15, when the virtual control line intersects with any triangular mesh in the 3D model corresponding to the intersecting bounding volume, it is determined that the virtual control line intersects with the 3D model corresponding to the intersecting bounding volume, so as to determine that a 3D Model.

In this step, in all the triangular meshes of the 3D model, as long as it is judged that the virtual control line intersects any one of the triangular meshes, it can be immediately judged that the virtual control line has detected a specific 3D model, and no longer Subsequent intersection detection is performed.

In the first embodiment of the present invention, the attitude angle information of the interactive pen is collected, and the rotation matrix of the interactive pen is determined according to the collected attitude angle of the interactive pen, and then according to the rotation matrix of the interactive pen, in the preset virtual scene The origin and the preset initial direction vector of the interactive pen map a virtual control line in the virtual scene in real time, and finally determine the bounding volume of the 3D model according to the stored vertex data information of the 3D model, and judge the relationship between the virtual control line and Whether the bounding volume of the 3D model intersects, judging whether the virtual control line intersects with any triangle mesh in the 3D model corresponding to the intersecting bounding volume, so as to determine whether the 3D model is detected. Since the bounding volume of the 3D model is detected first, after it is judged that the virtual control line intersects with the bounding volume of the 3D model, whether the virtual control line intersects with any triangular mesh in the 3D model is detected, thereby reducing the amount of computation. Therefore, the speed and accuracy of detecting the 3D model in the virtual environment scene are effectively improved.

Embodiment two:

FIG. 3 shows a structural diagram of a 3D model detection device in a virtual scene provided by a second embodiment of the present invention. For ease of description, only parts related to the embodiment of the present invention are shown.

The 3D model detection device in the virtual scene includes: a virtual control line generation unit 31, a model bounding volume construction unit 32, a model bounding volume intersection detection unit 33, a model triangle mesh intersection detection unit 34, a model detection and determination unit 35. in:

The virtual control line generating unit 31 is configured to map a virtual control line in the virtual scene in real time corresponding to the spatial attitude angle of the interactive pen.

Among them, the attitude angle of the interactive pen includes the pitch angle, roll angle and yaw angle of the interactive pen, and the attitude angle of the interactive pen is determined by the following method:

FIG. 4 shows a structural diagram of a 3D model detection device in a virtual scene provided by another embodiment of the present invention. At this time, the 3D model detection device in the virtual scene includes:

The initial pitch angle and roll angle determination unit 36 is configured to determine the initial pitch angle and roll angle of the interactive pen. Among them, the initial pitch angle and roll angle are determined by a three-axis gyroscope and a three-axis accelerometer installed on the interactive pen.

The final pitch angle and roll angle determination unit 37 is configured to iteratively determine the final pitch angle and roll angle according to the Kalman filter and the initial pitch angle and roll angle. The initial pitch angle and roll angle are represented by quaternion and used as the state quantity of the Kalman filter system. The attitude kinematics equation is used as the state transition equation of the Kalman filter, and the acceleration information is used as the observation information of the filter, and then the calculation method of the Kalman filter is used for iterative calculation to update the initial pitch angle and roll angle to an accurate and stable pitch angle and roll angle.

The magnetic field strength component determining unit 38 is used to determine the magnetic field strength component in the horizontal plane according to the final pitch angle and roll angle and the magnetic field strength components of the x-axis, y-axis, and z-axis of the coordinate system where the interactive pen is located.

The yaw angle determining unit 39 is used to determine the yaw angle according to the determined magnetic field intensity component in the horizontal plane.

Preferably, the virtual control line generating unit 31 includes:

The rotation matrix determining module 311 is configured to determine the rotation matrix of the interactive pen according to the attitude angle of the interactive pen. The rotation matrix of the interactive pen includes a matrix obtained by rotating the coordinate system of the interactive pen by an angle θ around the y-axis, where θ is the pitch angle; a matrix obtained by rotating the coordinate system of the interactive pen by an angle α around the z-axis, and the α is Yaw angle; the matrix obtained by rotating the coordinate system of the interactive pen around the x-axis by an angle of γ, where γ is the roll angle.

The virtual control line rendering module 312 is used to render the interactive pen in the virtual scene in real time according to the rotation matrix of the interactive pen, the preset origin in the virtual scene, and the preset initial direction vector of the interactive pen. Corresponding virtual control line. Suppose the initial direction vector of the preset interactive pen is After the interactive pen is rotated, the corresponding direction vector in the virtual scene (that is, the real-time mapped virtual control line) is R is a rotation matrix, then Where "·" represents the inner product operation.

The model enclosure construction unit 32 is configured to construct an enclosure corresponding to each 3D model in the virtual scene, the enclosure encloses the corresponding 3D model, and the 3D model is composed of a plurality of triangular meshes .

Among them, the bounding body of the 3D model is composed of the maximum vertex coordinates of the 3D model on the x-axis, the minimum vertex coordinates, the maximum vertex coordinates on the y-axis, the minimum vertex coordinates, the maximum vertex coordinates on the z-axis, and the minimum vertex coordinates. composition.

The model bounding volume intersection detection unit 33 is configured to determine whether the real-time mapped virtual control line intersects with the bounding volume of the 3D model.

Specifically, the intersection detection unit 33 of the model bounding volume includes:

The vector function determination module 331 is used to determine the function f(x, y, z) corresponding to the virtual control line, and the component of any point of f(x, y, z) on the x-axis, y-axis and z-axis is f _x (x,y,z), f _y (x,y,z), f _z (x,y,z).

Intersection condition judging module 332, for judging any point of f(x, y, z) on x-axis, y-axis, z-axis component f _x (x, y, z), f _y (x, y, z), Does f _z (x,y,z) all satisfy the following conditions:

min x ≤ f _x (x, y, z) ≤ max x;

min y≤f _y (x,y,z)≤max y;

min z ≤ f _z (x, y, z) ≤ max z.

Wherein, the min x, max x, min y, max y, min z, max z are the vertex coordinates of the bounding volume. Among them, min x, max x, min y, max y, min z, max z are the minimum and maximum vertex coordinates of the 3D model on the x-axis; the minimum and maximum vertex coordinates on the y-axis; The minimum and maximum vertex coordinates of .

The bounding volume intersection judging module 333 is used for the components f x (x, y, z) and f y (x, y, z) at any point of f ( _x , y, z) on the x-axis, y-axis, _and z-axis , f _z (x, y, z) all satisfy the above conditions, it is determined that the virtual control line intersects the bounding volume of the 3D model, otherwise, it is determined that the virtual control line does not intersect the bounding volume of the 3D model.

The intersection detection unit 34 of the model triangular mesh is used for judging any triangular mesh in the 3D model corresponding to the virtual control line and the intersecting bounding volume when the virtual control line intersects with the bounding volume of the 3D model Whether to intersect.

Specifically, the intersection detection unit 34 of the model triangular mesh includes:

Vertex weight value range judging module 341, used for when the parameter equation of the triangle corresponding to any triangle mesh in the 3D model is (1-uv)V ₀ +uV ₁ +vV ₂ , according to the origin in the preset virtual scene , the virtual control line and the three vertex coordinates of any triangular mesh in the 3D model, and judge whether u and v meet the following conditions: u≥0, v≥0, u+v≤1, wherein, (1- uv), u, v are the weights of the three vertices of the triangle respectively; V ₀ , V ₁ , V ₂ are the coordinates of the three vertices of the triangle.

The triangular mesh intersection judging module 342 is used to judge the virtual control line in the 3D model corresponding to the intersecting bounding volume when u and v satisfy u≥0, v≥0, and u+v≤1. Any triangular mesh intersects, when u and v do not satisfy u≥0, v≥0, u+v≤1, it is determined that the virtual control line corresponds to any triangle in the 3D model corresponding to the intersecting bounding volume Meshes are disjoint.

The specific detection process is the same as that in Embodiment 1, and will not be repeated here.

A model detection and determination unit 35, configured to determine that the virtual control line intersects with the 3D model corresponding to the intersecting volume when the virtual control line intersects with any triangle mesh in the 3D model corresponding to the intersecting volume, A 3D model is detected with a judgment.

In the second embodiment of the present invention, since the bounding volume of the 3D model is detected first, the virtual control line and any triangular mesh in the 3D model are detected after it is judged that the virtual control line intersects with the bounding volume of the 3D model Whether or not they intersect, thus reducing the amount of computation, thereby effectively improving the speed and accuracy of detecting 3D models in the virtual environment scene.

Those of ordinary skill in the art can understand that all or part of the steps in the method of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. Storage media, such as ROM/RAM, magnetic disk, optical disk, etc.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. within range.

Claims (10)

1. the 3D model checking method in virtual scene, is characterized in that, described method comprises the steps:

The spatial attitude angle of corresponding described interaction pen maps out in real time a virtual controlling line in described virtual scene;

Each 3D model in corresponding described virtual scene, constructs respectively an enclosure body, and described enclosure body wraps corresponding 3D model, and described 3D model is comprised of a plurality of triangle griddings;

Judge whether the virtual controlling line of described real-time mapping and the enclosure body of 3D model intersect;

When the enclosure body of described virtual controlling line and 3D model intersects, judge whether the arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects;

When the arbitrary triangle gridding in the described virtual controlling line 3D model corresponding with crossing enclosure body intersects, judge that the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects, and detects 3D model to judge.

2. the method for claim 1, is characterized in that, described spatial attitude angle comprises the angle of pitch, roll angle and crab angle; Spatial attitude angle at the described interaction pen of described correspondence, before mapping out in real time the step of a virtual controlling line, comprises the steps: in described virtual scene

Determine the angle of pitch and roll angle that interaction pen is initial;

According to Kalman filtering and the described initial angle of pitch and roll angle iteration, determine the final angle of pitch and roll angle;

According to the magnetic field strength component of the x axle of the described final angle of pitch and roll angle and interaction pen place coordinate system, y axle, z axle, determine the magnetic field strength component in surface level;

According to the magnetic field strength component in definite surface level, determine crab angle.

3. the method for claim 1, is characterized in that, the spatial attitude angle of the described interaction pen of described correspondence, and the step that maps out in real time a virtual controlling line in described virtual scene specifically comprises:

According to the attitude angle of described interaction pen, determine the rotation matrix of described interaction pen;

According to the inceptive direction vector of the initial point in the rotation matrix of described interaction pen, default virtual scene and default interaction pen, in virtual scene, real-time rendering goes out corresponding virtual controlling line after described interaction pen rotation.

4. the method for claim 1, is characterized in that, describedly judges that the virtual controlling line of described real-time mapping and the whether crossing step of the enclosure body of 3D model specifically comprise:

Determine the function f (x, y, z) that described virtual controlling line is corresponding, f (x, y, z) any point is f at the component of x axle, y axle, z axle _x(x, y, z), f _y(x, y, z), f _z(x, y, z);

Judgement f (x, y, z) any point is at the component f of x axle, y axle, z axle _x(x, y, z), f _y(x, y, z), f _zwhether (x, y, z) all meets the following conditions:

min?x≤f _x(x,y,z)≤max?x；

min?y≤f _y(x,y,z)≤max?y；

min?z≤f _z(x,y,z)≤max?z；

Wherein, the apex coordinate that described min x, max x, min y, max y, min z, max z are described enclosure body;

Component f at f (x, y, z) any point at x axle, y axle, z axle _x(x, y, z), f _y(x, y, z), f _zwhen (x, y, z) all meets above-mentioned condition, judge that the enclosure body of described virtual controlling line and 3D model intersects, otherwise, judge that the enclosure body of described virtual controlling line and 3D model is non-intersect.

5. the method for claim 1, it is characterized in that, when described arbitrary triangle gridding in the described virtual controlling line 3D model corresponding with crossing enclosure body intersects, judge that the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects, the step that 3D model detected to judge specifically comprises:

The leg-of-mutton parametric equation that in 3D model, arbitrary triangle gridding is corresponding is (1-u-v) V ₀+ uV ₁+ vV ₂, wherein, (1-u-v), u, v be respectively the weight on an Atria summit; V ₀, V ₁, V ₂coordinate for an Atria summit;

According to three apex coordinates of the arbitrary triangle gridding in the initial point in default virtual scene, described virtual controlling line and 3D model, judge whether u, v meet the following conditions: u >=0, v >=0, u+v≤1, if meet, judge that the described arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects, otherwise, judge that the described arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body is non-intersect.

6. the 3D model pick-up unit in virtual scene, is characterized in that, described device comprises:

Virtual controlling line generation unit for the spatial attitude angle of the described interaction pen of correspondence, maps out in real time a virtual controlling line in described virtual scene;

Model enclosure body tectonic element, each the 3D model for the described virtual scene of correspondence, constructs respectively an enclosure body, and described enclosure body wraps corresponding 3D model, and described 3D model is comprised of a plurality of triangle griddings;

The Intersection detecting unit of model enclosure body, for judging whether the virtual controlling line of described real-time mapping and the enclosure body of 3D model intersect;

The Intersection detecting unit of model triangle gridding, while intersecting for the enclosure body at described virtual controlling line and 3D model, judges whether the arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects;

Model detects identifying unit, while intersecting for the arbitrary triangle gridding at the described virtual controlling line 3D model corresponding with crossing enclosure body, judges that the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects, and detects 3D model to judge.

7. device as claimed in claim 6, is characterized in that, described device comprises:

The initial angle of pitch and roll angle determining unit, for determining the initial angle of pitch and the roll angle of interaction pen;

The final angle of pitch and roll angle determining unit, for determining the final angle of pitch and roll angle according to Kalman filtering and the described initial angle of pitch and roll angle iteration;

Magnetic field strength component determining unit, for according to the magnetic field strength component of the x axle of the described final angle of pitch and roll angle and interaction pen place coordinate system, y axle, z axle, determines the magnetic field strength component in surface level;

Crab angle determining unit, for determining crab angle according to the magnetic field strength component in definite surface level.

8. device as claimed in claim 6, is characterized in that, described virtual controlling line generation unit comprises:

Rotation matrix determination module, for determining the rotation matrix of described interaction pen according to the attitude angle of described interaction pen;

Virtual controlling line rendering module, for according to the inceptive direction vector of the initial point of the rotation matrix of described interaction pen, default virtual scene and default interaction pen, in virtual scene, real-time rendering goes out corresponding virtual controlling line after described interaction pen rotation.

9. device as claimed in claim 6, is characterized in that, the Intersection detecting unit unit of described model enclosure body comprises:

Vector function determination module, for determining function f (x, y, z) corresponding to described virtual controlling line, f (x, y, z) any point is f at the component of x axle, y axle, z axle _x(x, y, z), f _y(x, y, z), f _z(x, y, z);

Intersect condition judgment module, for judging that f (x, y, z) any point is at the component f of x axle, y axle, z axle _x(x, y, z), f _y(x, y, z), f _zwhether (x, y, z) all meets the following conditions:

min?x≤f _x(x,y,z)≤max?x；

min?y≤f _y(x,y,z)≤max?y；

min?z≤f _z(x,y,z)≤max?z；

Wherein, the apex coordinate that described min x, max x, min y, max y, min z, max z are described enclosure body;

Enclosure body intersects judge module, for the component f at x axle, y axle, z axle at f (x, y, z) any point _x(x, y, z), f _y(x, y, z), f _zwhen (x, y, z) all meets above-mentioned condition, judge that the enclosure body of described virtual controlling line and 3D model intersects, otherwise, judge that the enclosure body of described virtual controlling line and 3D model is non-intersect.

10. device as claimed in claim 6, is characterized in that, the Intersection detecting unit of described model triangle gridding comprises:

Vertex weights value scope judge module, for being (1-u-v) V at leg-of-mutton parametric equation corresponding to the arbitrary triangle gridding of 3D model ₀+ uV ₁+ vV ₂time, according to three apex coordinates of the arbitrary triangle gridding in the initial point in default virtual scene, described virtual controlling line and 3D model, judge whether u, v meet the following conditions: u>=0, v>=0, u+v≤1, wherein, (1-u-v), u, v are respectively the weight on an Atria summit; V ₀, V ₁, V ₂coordinate for an Atria summit;

Triangle gridding Intersection determination module, for meeting u >=0 at u, v, v >=0, u+v≤1 o'clock, judge that the described arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body intersects, and does not meet u >=0, v >=0 at u, v, u+v≤1 o'clock, judges that the described arbitrary triangle gridding in the 3D model that described virtual controlling line is corresponding with crossing enclosure body is non-intersect.