A three-dimensional model construction method, device, system and storage medium

Abstract

The embodiment of the present invention provides a method, device, system and storage medium for constructing a three-dimensional model. The construction method includes: obtaining mesh information and skeleton information of the head and torso of a three-dimensional human body model; determining key points of the head and torso according to the mesh information and the skeleton information; constructing a head proxy body according to the head key points, and constructing a torso proxy body according to the torso key points; replacing the head and torso of the three-dimensional human body model with the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional model technology, and in particular to a three-dimensional model construction method, device, system and storage medium. Background Art

[0002] 3D virtual characters have many applications in animation production and movie special effects. The general process is to capture the various movements of the actors through motion capture technology, and then transfer these movements to the specified 3D model.

[0003] However, there will be many problems during the migration process. For example, due to the different body shapes of the actors and the 3D models to be driven, the 3D model may make some movements that are not in line with the principles of physics, such as reaching the hand into the inside of the model. Summary of the invention

[0004] In order to solve the problems existing in the prior art, at least one embodiment of the present invention provides a three-dimensional model construction method, device, system and storage medium.

[0005] In a first aspect, an embodiment of the present invention provides a three-dimensional model construction method, the construction method comprising:

[0006] Obtain mesh information and bone information of the head and torso of the human body 3D model;

[0007] Determine the key points of the head and the key points of the torso according to the mesh information and the skeleton information;

[0008] A head proxy body is constructed according to the head key points, and a torso proxy body is constructed according to the torso key points;

[0009] The head and torso of the three-dimensional human body model are replaced by the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

[0010] Based on the above technical solution, the embodiment of the present invention can also make the following improvements.

[0011] In combination with the first aspect, in a first embodiment of the first aspect, the key points of the head include: grid points corresponding to the eyes, nose tip, mouth corners, chin tip, ears, top of the head, head circumference and back of the head, respectively.

[0012] In combination with the first embodiment of the first aspect, in a second embodiment of the first aspect, a method for determining grid points corresponding to a head circumference includes:

[0013] Acquire three-dimensional information of neck joints according to the bone information of the head;

[0014] The average of the vertical coordinates of the grid points corresponding to the top of the head and the tip of the nose is used as the vertical coordinate of the head circumference;

[0015] Determine, according to the three-dimensional information of the neck joint point and the grid information of the grid point corresponding to the nose tip, the neck projection point and the nose tip projection point of the neck joint point and the grid point corresponding to the nose tip on the cross section in the horizontal direction of the head circumference;

[0016] Taking the neck projection point as the origin, four rays are generated on the cross section in the horizontal direction where the head circumference is located; wherein the four rays divide the cross section in the horizontal direction where the head circumference is located into four equal parts, and the line connecting the neck projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point;

[0017] According to the grid information of the head, the grid point closest to the head circumference intersection is used as the grid point corresponding to the head circumference; wherein the head circumference intersection is the intersection of the ray and the grid structure of the head.

[0018] In combination with the first embodiment of the first aspect, in a third embodiment of the first aspect, a method for determining a grid point corresponding to the back of the head includes:

[0019] Acquire three-dimensional information of neck joints according to the bone information of the head;

[0020] Determine, according to the three-dimensional information of the neck joint point, a neck projection point of the neck joint point on a cross section in a horizontal direction where a grid point corresponding to the nose tip is located;

[0021] According to the grid information of the head, the grid point closest to the intersection of the back of the head is used as the grid point corresponding to the back of the head; wherein the intersection of the back of the head is the intersection of the extended line of the line connecting the grid point corresponding to the nose tip and the neck projection point and the grid structure of the head.

[0022] In combination with the first embodiment of the first aspect, in a fourth embodiment of the first aspect, the method for determining the key points of the torso includes:

[0023] Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, and the three-dimensional information of the second spine joint according to the skeleton information of the trunk;

[0024] According to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point, determine the average of the coordinates of the left shoulder joint point and the right shoulder joint point in the vertical direction as the coordinate of the trunk key point in the vertical direction;

[0025] Determine, based on the three-dimensional information of the second spinal joint point and the grid information of the grid point corresponding to the nose tip, the second spinal projection point and the nose tip projection point of the second spinal joint point and the grid point corresponding to the nose tip on the cross section in the horizontal direction where the trunk key point is located;

[0026] Determine, according to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point, a left shoulder projection point and a right shoulder projection point of the left shoulder joint point and the right shoulder joint point on a cross section in a horizontal direction where the key point of the trunk is located;

[0027] Taking the second spine projection point as the starting point, and according to the second spine projection point and the nose tip projection point, establishing a first ray from the second spine projection point to the nose tip projection point;

[0028] According to the mesh information of the torso, the mesh point closest to the torso intersection is used as the torso key point; wherein the torso intersection is the intersection of a second ray constructed with the left shoulder projection point as the starting point and the mesh structure of the torso, and the torso intersection is the intersection of a third ray constructed with the right shoulder projection point as the starting point and the mesh structure of the torso, and the directions of the first ray, the second ray and the third ray are the same.

[0029] In combination with the first embodiment of the first aspect, in a fifth embodiment of the first aspect, the method for determining the key points of the torso includes:

[0030] Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spinal joint, and the three-dimensional information of the second spinal joint according to the skeleton information of the trunk; wherein the coordinate of the first spinal joint in the vertical direction is greater than the coordinate of the second spinal joint in the vertical direction;

[0031] Establishing a cross section in the horizontal direction of the first spinal joint point as a spinal reference cross section;

[0032] Determine a left shoulder projection point and a right shoulder projection point of the left shoulder joint point and the right shoulder joint point on the spine reference section according to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point;

[0033] Establishing a line connecting the left shoulder projection point and the right shoulder projection point as a reference line;

[0034] Establishing a vertical line with the first spinal joint point as the orthocenter to divide the spinal reference section into a first reference section and a second reference section; wherein the vertical line is perpendicular to the reference line;

[0035] Move both the first reference section and the second reference section vertically downward to the second spinal joint point, and obtain the distance between each grid point in the first reference section and the reference line in real time according to the grid information of the trunk, and obtain the distance between each grid point in the second reference section and the reference line in real time;

[0036] The grid point with the largest distance from the reference line in the first reference section and the grid point with the largest distance from the reference line in the second reference section are used as the key points of the torso.

[0037] In combination with the first embodiment of the first aspect, in a sixth embodiment of the first aspect, the method for determining the key points of the torso includes:

[0038] Determine the three-dimensional information of the hip joint and the three-dimensional information of the first spine joint according to the skeleton information of the trunk;

[0039] Determine, based on the three-dimensional information of the first spinal joint point and the grid information of the grid points corresponding to the nose tip, a first spinal projection point and a nose tip projection point of the grid points corresponding to the first spinal joint point and the nose tip on a cross section in the horizontal direction where the hip joint point is located;

[0040] Taking the first spine projection point as the origin, three rays are generated on the cross section in the horizontal direction where the hip joint point is located; wherein the three rays divide the cross section in the horizontal direction where the hip joint point is located into three equal parts, and the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point;

[0041] According to the mesh information of the torso, the mesh point closest to the hip intersection is used as the key point of the torso; wherein the hip intersection is the intersection point of the ray and the mesh structure of the torso.

[0042] In combination with the first embodiment of the first aspect, in a seventh embodiment of the first aspect, the method for determining the key points of the torso includes:

[0043] Determine the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point according to the skeleton information of the torso;

[0044] According to the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point, determine the average of the coordinates of the left hip joint point and the right hip joint point in the vertical direction as the coordinate of the trunk key point in the vertical direction;

[0045] Establishing a cross section in the horizontal direction at the key point of the torso as a reference cross section of the hip bone;

[0046] The three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point are used to determine the left hip joint point and the left hip joint point and the right hip joint point on the hip reference section;

[0047] A first ray and a second ray located on the hip bone reference section are established with the left hip bone projection point as a starting point; wherein a line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the first ray and the second ray;

[0048] A third ray and a fourth ray located on the hip bone reference section are established with the right hip bone projection point as the starting point; wherein the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the third ray and the fourth ray;

[0049] According to the grid information of the torso, the grid point closest to the hip intersection is used as the key point of the torso; wherein the hip intersection is the intersection of the first ray, the second ray, the third ray and the fourth ray with the grid structure of the torso respectively.

[0050] In combination with the first aspect or the first, second, third, fourth, fifth, sixth or seventh embodiment of the first aspect, in an eighth embodiment of the first aspect, constructing a head proxy body according to the head key points, and constructing a torso proxy body according to the torso key points, comprises:

[0051] Connecting the head key points in pairs to obtain a first grid structure;

[0052] Removing the lines inside the first grid structure to obtain a head grid structure as the head proxy body;

[0053] Connecting the trunk key points in pairs to obtain a second mesh structure;

[0054] The connecting lines inside the second grid structure are removed to obtain a trunk grid structure as the trunk proxy body.

[0055] In a second aspect, an embodiment of the present invention provides a three-dimensional model construction device, the construction device comprising:

[0056] A first processing unit is used to obtain mesh information and bone information of a head and a torso of a three-dimensional human body model;

[0057] A second processing unit, configured to determine a head key point and a torso key point according to the mesh information and the skeleton information;

[0058] A third processing unit, configured to construct a head proxy body according to the head key points, and to construct a torso proxy body according to the torso key points;

[0059] The fourth processing unit is used to replace the head and the torso of the three-dimensional human body model with the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

[0060] In a third aspect, an embodiment of the present invention provides a three-dimensional model construction system, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0061] Memory, used to store computer programs;

[0062] The processor is used to implement the three-dimensional model construction method described in any embodiment of the first aspect when executing the program stored in the memory.

[0063] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the three-dimensional model construction method described in any embodiment of the first aspect.

[0064] The above technical solution of the present invention has the following advantages compared with the prior art: the embodiment of the present invention obtains the mesh information and skeleton information of the head and torso of the human body three-dimensional model, determines the key points of the head and torso according to the mesh information and the skeleton information, constructs proxy bodies corresponding to the head and torso according to the key points of the head and torso, replaces the corresponding parts in the human body three-dimensional model with the proxy bodies, so as to simplify the human body three-dimensional model, completes the motion redirection, collision calculation, penetration repair and other algorithms of low-precision scenes through the simplified human body three-dimensional model, avoids the above-mentioned calculation detection with the original human body three-dimensional model, thereby improving the calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of a system provided for each embodiment of the present invention;

[0066] Figure 2 It is a schematic diagram of the method flow provided by an embodiment of the present invention;

[0067] Figure 3 is a schematic diagram of a method flow chart provided by another embodiment of the present invention;

[0068] Figure 4 This is a schematic diagram of a method flow chart provided by another embodiment of the present invention;

[0069] Figure 5 This is a second schematic diagram of a method flow chart provided by another embodiment of the present invention;

[0070] Figure 6It is the third schematic diagram of the method flow provided by another embodiment of the present invention;

[0071] Figure 7 It is a third schematic diagram of a method flow chart provided by another embodiment of the present invention;

[0072] Figure 8 It is the third schematic diagram of the method flow provided by another embodiment of the present invention;

[0073] Fig. 9 It is the third schematic diagram of the method flow provided by another embodiment of the present invention;

[0074] Fig.10 is a schematic diagram of a device structure provided by another embodiment of the present invention;

[0075] Fig.11 This is a schematic diagram of a system structure provided by another embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0077] like Figure 1 As shown, it is a structural schematic diagram of a system for implementing various embodiments of the present invention, including a terminal, the terminal including: a processor 1110 and a memory 1130, wherein the terminal can be a mobile terminal such as a mobile phone, a tablet computer, a laptop computer, a handheld computer, a personal digital assistant (Personal Digital Assistant, PDA), a portable media player (Portable Media Player, PMP), a navigation device, a wearable device, a smart bracelet, a pedometer, etc., as well as a fixed terminal such as a digital TV, a desktop computer, etc.

[0078] like Figure 2 As shown, a three-dimensional model construction method is provided in an embodiment of the present invention. Figure 2 , the construction method includes the following steps:

[0079] S11. Obtain mesh information and bone information of the head and torso of the three-dimensional human body model.

[0080] In this embodiment, 3D human body modeling, as a component of computer human body simulation, has always been one of the hot topics of research. Since the birth of interactive computer graphics, scholars have been constantly exploring computer human body modeling technology. From wireframe modeling, solid modeling, surface modeling to physics-based modeling, significant progress has been made.

[0081] In this embodiment, with the development of 3D games and virtual reality technology, the requirements for the simulation degree of objects in the scene are getting higher and higher, and the requirements for the three-dimensional model are also increasing. Among them, since the three-dimensional human body model is generally controlled by the user, the user completes the complex actions in the scene by controlling the three-dimensional human body model, and the requirements for the three-dimensional human body model are getting higher and higher. Once the three-dimensional human body model fails to restore the action, or the limbs of the three-dimensional human body model overlap, this situation is called "model penetration" in the industry, which will seriously affect the user's experience. Therefore, after the three-dimensional human body model is established, the movement process of the three-dimensional human body model will be verified according to the operation logic of the three-dimensional human body model to confirm whether the three-dimensional human body model will have model penetration. Since the complete three-dimensional human body model is relatively delicate and complex, the system calculation amount of controlling the three-dimensional human body model to perform actions is very time-consuming.

[0082] In this embodiment, the human body 3D model includes the static 3D digital human body used in current industrial animation and games, which is mainly a 3D mesh model (generally called mesh in English). The number of faces of the mesh determines the fineness of the model itself. However, in games, since real-time rendering of a large number of meshes consumes computer performance very much, the number of meshes in actual games will not be too high. In practice, a model with a lower number of faces is used, and then a texture is attached to make it look very real.

[0083] In this embodiment, the mesh of the general human body surface is composed of triangular meshes to form the smallest surface, and the smaller the spacing between the mesh points constituting the mesh, the smaller the mesh, and the higher the accuracy of the final model. In this step, the mesh information of the head and torso is obtained, that is, the information of each mesh point of the human head and torso is obtained, such as the position information of the mesh points.

[0084] In this embodiment, the grid information refers to the position information of each grid point constituting the head and torso of the three-dimensional model of the human body.

[0085] In this embodiment, in existing projects that use three-dimensional human body models, when simulating the movement of the three-dimensional human body model, in order to improve the simulation effect of the three-dimensional human body model, a 3D model of the human skeleton is constructed for the three-dimensional human body model. The 3D model of the human skeleton is used to further determine the movement of the three-dimensional human body model. According to the movement of the bones, the movement of the grid points on the surface of the model is determined to improve the motion control accuracy of the three-dimensional model.

[0086] In this embodiment, the skeleton information refers to the skeleton point position information of the head and torso constituting the three-dimensional model of the human body.

[0087] In this embodiment, after obtaining the mesh information of the head and torso of the human body 3D model, the bone information of the head and torso of the human body 3D model is obtained. The bone information of the head and torso of the human body 3D model can be obtained through the human body skeleton 3D model.

[0088] S12. Determine the key points of the head and the torso according to the mesh information and the skeleton information.

[0089] In this embodiment, the key points of the head and the key points of the trunk are determined based on the mesh information and the skeleton information of the head and the trunk. Since the three-dimensional model needs to be simplified, in this step, the key points of the head and the trunk are obtained, which may be the points with more prominent shapes of the head and the trunk. In addition, in order to ensure that the simplified three-dimensional model can complete collision detection or penetration detection, in addition to determining the contour points on the head, points corresponding to more important areas may also be retained. For example, the key points of the head may include: points corresponding to the eyes, nose tip, corners of the mouth, chin tip, ears, top of the head, head circumference and back of the head, and the key points of the trunk may be determined synchronously based on the skeleton information and the mesh information. For example, the key points of the trunk may include: contour points corresponding to the shoulders, contour points corresponding to the chest, contour points corresponding to the abdomen and contour points corresponding to the hips. However, due to the different simplicity of the three-dimensional model, we may also use the mesh points closest to the position as the corresponding key points after confirming the accurate position to reduce the amount of calculation.

[0090] S13. Construct a head proxy body according to the key points of the head, and construct a torso proxy body according to the key points of the torso.

[0091] In this embodiment, the corresponding head proxy body and torso proxy body are constructed according to the head key points and the torso key points respectively; for example, the head proxy body is constructed according to the head key points, and a triangular mesh can be constructed with the head key points to form the head proxy body, ensuring that no head key points are inside the head proxy body, or making sure that no triangular mesh is not inside the head proxy body, then the construction of the head proxy body is completed, and similarly, the construction of the torso proxy body can be completed.

[0092] In this embodiment, since the head key points and the torso key points are partial points in the original human three-dimensional model, the proxy body constructed with the head key points and the torso key points is simpler than the original three-dimensional model. However, since the simplified model needs to be able to complete the motion redirection, collision calculation, and mold penetration repair algorithms of the original human three-dimensional model, once the selected key points deviate, the final simplified proxy body will be unable to replace the original three-dimensional model, and the above-mentioned problem cannot be tested through the proxy body.

[0093] S14. Replace the head and torso of the three-dimensional human body model with the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

[0094] In this embodiment, the key points of the head and the torso are extracted according to the mesh information and the skeleton information of the original model, and a proxy body is constructed corresponding to the key points of the head and the torso, so as to realize the proxy of the three-dimensional human body model and obtain a simplified three-dimensional human body model. Since the proxy body only contains key points compared to the original three-dimensional human body model, when the simplified three-dimensional human body model is used for relevant detection, the data operation of the simplified model can be quickly realized.

[0095] In this embodiment, this scheme obtains the mesh information and bone information of the head and torso of the human body three-dimensional model, determines the key points of the head and torso according to the mesh information and the bone information, constructs proxy bodies corresponding to the head and torso according to the key points of the head and torso, and replaces the corresponding parts in the human body three-dimensional model with the proxy bodies to simplify the human body three-dimensional model. The simplified human body three-dimensional model is used to complete the motion redirection, collision calculation, penetration repair and other algorithms of low-precision scenes, avoids the above-mentioned calculation detection with the original human body three-dimensional model, thereby improving the calculation efficiency.

[0096] like Figure 3 As shown, in this embodiment, in S14, a head proxy body is constructed according to the key points of the head, and a torso proxy body is constructed according to the key points of the torso, including the following steps:

[0097] S21. Connect the key points of the head in pairs to obtain a first grid structure.

[0098] S22: Remove the internal connection lines of the first grid structure to obtain a head grid structure as a head proxy body.

[0099] In this embodiment, in order to speed up the construction of the head proxy body, in this scheme, by connecting the key points of the head in pairs, a grid structure with internal connections will eventually be obtained. At this time, all the connections in the grid structure are removed to obtain a proxy body composed of triangular meshes similar to the original human body three-dimensional model. Since the number of head key points is greatly reduced compared to the number of grid points of the human body three-dimensional model, the triangular mesh of the head proxy body finally formed will be larger and rougher than the original human body three-dimensional model.

[0100] S23. Connect the trunk key points in pairs to obtain a second mesh structure.

[0101] S24. Remove the connection lines inside the second grid structure to obtain a torso grid structure as a torso proxy body.

[0102] In this embodiment, in order to speed up the construction of the torso proxy body, in this scheme, by connecting the torso key points in pairs, a grid structure with internal connections will eventually be obtained. At this time, all the connections in the grid structure are removed to obtain a proxy body composed of triangular meshes similar to the original human three-dimensional model. Since the number of torso key points is greatly reduced compared to the number of grid points of the human three-dimensional model, the triangular mesh of the torso proxy body finally formed will be larger and rougher than the original human three-dimensional model.

[0103] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining grid points corresponding to head circumference.

[0104] like Figure 4 Specifically, in this embodiment, the method for determining the grid points corresponding to the head circumference includes the following steps:

[0105] S31. Acquire three-dimensional information of neck joints based on the head bone information.

[0106] In this embodiment, the three-dimensional space we usually refer to is a space system formed by adding a direction vector to the two-dimensional plane system. The three-dimensional space is the three axes of the coordinate system, namely the x-axis, the y-axis, and the z-axis, where x represents the left and right space, y represents the front and back space, and z represents the up and down space (the plane rectangular coordinate system cannot be used to understand the spatial direction).

[0107] In this embodiment, the three-dimensional information refers to three-dimensional position information.

[0108] In this embodiment, the three-dimensional information of the neck joint point in the skeleton information is obtained through the skeleton information of the head. Since the neck supports the head in the human body structure, the neck is equivalent to the center of gravity of the head, and the head circumference is a measure of the size of the head, which is closely related to the establishment of the head proxy body. In this solution, the neck joint point can be used as the center point of the head.

[0109] S32. Taking the average of the vertical coordinates of the grid points corresponding to the top of the head and the tip of the nose as the vertical coordinates of the head circumference.

[0110] In this embodiment, in normal head circumference measurement, the head circumference is measured through the protrusion above the eyebrow arch and around the occipital tubercle. However, the above points are not set in detail in the three-dimensional human body model. Therefore, in this step, the average value of the vertical coordinates of the grid points corresponding to the top of the head and the tip of the nose in the three-dimensional human body model is used as the vertical coordinates of the plane where the head circumference is located, so as to facilitate the subsequent determination of the horizontal section passing through the head circumference.

[0111] S33. Determine the neck projection points and nose tip projection points of the neck joint points and the nose tip corresponding grid points on the cross section in the horizontal direction of the head circumference according to the three-dimensional information of the neck joint points and the grid information of the nose tip corresponding grid points.

[0112] In this embodiment, based on the three-dimensional information of the neck joint points and the grid information of the grid points corresponding to the nose tip, the neck projection points and the nose tip projection points of the neck joint points and the grid points corresponding to the nose tip on the horizontal cross-section of the head circumference are obtained. That is, based on the three-dimensional information of the neck joint points, the coordinates of the neck joint points in the vertical direction are replaced with the coordinates of the horizontal cross-section in the vertical direction to obtain the coordinates of the neck projection points. Similarly, the coordinates of the nose tip projection points can be obtained.

[0113] S34. With the neck projection point as the origin, four rays are generated on the cross section in the horizontal direction where the head circumference is located; wherein the four rays divide the cross section in the horizontal direction where the head circumference is located into four equal parts, and the line connecting the neck projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point.

[0114] In this embodiment, with the neck joint point as the starting point, four rays in the cross section in the horizontal direction of the head circumference are generated, the horizontal cross section is divided into four equal parts by the rays, and it is ensured that the line connecting the neck projection point and the nose tip projection point bisects the angle formed by two of the rays.

[0115] Specifically, the projection of the nose tip point on the cross section is connected to the neck projection point and the nose tip projection point. Then, the reference direction is from the neck projection point to the nose tip projection point. The reference direction is 0 degrees and the neck projection point is the origin. Four rays are emitted with angles of 45 degrees, 135 degrees, 225 degrees and 315 degrees, respectively. The four rays described in the above embodiment can be obtained.

[0116] S35. According to the grid information of the head, the grid point closest to the head circumference intersection is used as the grid point corresponding to the head circumference; wherein the head circumference intersection is the intersection of the ray and the grid structure of the head.

[0117] In this embodiment, according to the grid information of the head, the intersection point of the ray and the grid structure of the head is used as the head circumference intersection point, and the grid point closest to the head circumference intersection point is used as the grid point corresponding to the head circumference.

[0118] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining the grid points corresponding to the back of the head.

[0119] like Figure 5 As shown, specifically, in this embodiment, the method for determining the grid point corresponding to the back of the head includes the following steps:

[0120] S41. Acquire three-dimensional information of neck joints based on the head bone information.

[0121] In this embodiment, the three-dimensional information of the neck joint point in the skeleton information is obtained through the skeleton information of the head. Since the neck supports the head in the human body structure, the neck is equivalent to the center of gravity of the head, and the head circumference is a measure of the size of the head, which is closely related to the establishment of the head proxy body. In this solution, the neck joint point can be used as the center point of the head.

[0122] S42. Determine the neck projection point of the neck joint point on the cross section in the horizontal direction where the grid point corresponding to the nose tip is located according to the three-dimensional information of the neck joint point.

[0123] In this embodiment, based on the grid information of the three-dimensional information of the neck joint point, the neck projection point of the neck joint point on the horizontal cross-section where the grid point corresponding to the nose tip is located is obtained, that is, based on the three-dimensional information of the neck joint point, the coordinates of the neck joint point in the vertical direction are replaced with the coordinates of the horizontal cross-section in the vertical direction to obtain the coordinates of the neck projection point.

[0124] S43. According to the grid information of the head, the grid point closest to the intersection point of the back of the head is used as the grid point corresponding to the back of the head; wherein the intersection point of the back of the head is the intersection point of the extended line of the line connecting the grid point corresponding to the nose tip and the projection point of the neck and the grid structure of the head.

[0125] In this embodiment, according to the grid information of the head, the intersection point of the extended line of the line connecting the grid point corresponding to the nose tip and the neck projection point and the grid structure of the head is taken as the intersection point of the back of the head, and the grid point closest to the intersection point of the back of the head is taken as the grid point corresponding to the head circumference.

[0126] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining the key points of the torso.

[0127] like Figure 6 Specifically, in this embodiment, the method for determining the key points of the torso includes the following steps:

[0128] S51. Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, and the three-dimensional information of the second spine joint according to the skeleton information of the trunk.

[0129] In this embodiment, the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint and the three-dimensional information of the second spine joint are obtained through the skeleton information of the torso. Since the spine is the center of the torso in the human body structure, the shoulder key points of the torso part of the human body three-dimensional model should be the grid points that reflect the position of the shoulder thickness, and the shoulder joint points on the torso are the points corresponding to the thicker shoulder areas.

[0130] S52. According to the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, determine the average of the vertical coordinates of the left shoulder joint and the right shoulder joint as the vertical coordinate of the trunk key point.

[0131] In this embodiment, the average of the vertical coordinates of the left shoulder joint point and the right shoulder joint point is used as the vertical coordinate of the trunk joint point, wherein the trunk key point obtained in this embodiment is the trunk key point corresponding to the shoulder.

[0132] S53. Determine the second spine projection point and the nose tip projection point of the second spine joint point and the nose tip corresponding grid point on the horizontal cross section where the trunk key point is located based on the three-dimensional information of the second spine joint point and the grid information of the nose tip corresponding grid point.

[0133] In this embodiment, based on the three-dimensional information of the second spinal joint point and the grid information of the grid points corresponding to the nose tip, the second spinal projection points and the nose tip projection points are obtained on the horizontal cross-section where the trunk key points are located. That is, based on the three-dimensional information of the second spinal joint point, the coordinates of the second spinal joint point in the vertical direction are replaced with the coordinates of the horizontal cross-section in the vertical direction to obtain the coordinates of the second spinal projection point. Similarly, the coordinates of the nose tip projection point can be obtained.

[0134] S54. Determine the left shoulder projection point and the right shoulder projection point of the left shoulder joint and the right shoulder joint on the cross section in the horizontal direction where the trunk key point is located according to the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint.

[0135] In this embodiment, in this embodiment, based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the left shoulder projection point and the right shoulder projection point on the horizontal cross-section where the key point of the torso is located are obtained, that is, based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the coordinates of the left shoulder joint in the vertical direction are replaced with the coordinates of the horizontal cross-section in the vertical direction to obtain the coordinates of the left shoulder projection point, and similarly, the coordinates of the right shoulder projection point can be obtained.

[0136] S55. Taking the second spine projection point as the starting point, and based on the second spine projection point and the nose tip projection point, establish a first ray from the second spine projection point to the nose tip projection point.

[0137] In this embodiment, the second spine projection point is taken as the starting point, and a first ray is established from the second spine projection point to the nose tip projection point. The first ray can be used to determine the front of the human body three-dimensional model.

[0138] S56. According to the grid information of the torso, the grid point closest to the torso intersection is used as the key point of the torso; wherein the torso intersection is the intersection of the second ray constructed with the left shoulder projection point as the starting point and the grid structure of the torso, and the torso intersection is the intersection of the third ray constructed with the right shoulder projection point as the starting point and the grid structure of the torso, and the directions of the first ray, the second ray and the third ray are the same.

[0139] In this embodiment, the left shoulder projection point and the right shoulder projection point are used as starting points respectively to construct the second ray and the third ray, ensuring that the directions of the first ray, the second ray and the third ray are the same. Then the second ray and the third ray both point to the front of the three-dimensional model of the human body. The intersection of the second ray and the third ray with the grid structure of the torso is the projection of the left shoulder projection point and the right shoulder projection point in front of the torso. The grid point closest to the intersection of the torso is used as the key point of the torso. The above embodiment is to confirm the shoulder contour point of the corresponding shoulder on the torso.

[0140] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining the key points of the torso.

[0141] like Figure 7 Specifically, in this embodiment, the method for determining the key points of the torso includes the following steps:

[0142] S61. Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spinal joint and the three-dimensional information of the second spinal joint based on the skeletal information of the trunk; wherein the coordinate of the first spinal joint in the vertical direction is greater than the coordinate of the second spinal joint in the vertical direction.

[0143] In this embodiment, the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spine joint and the three-dimensional information of the second spine joint are obtained through the skeletal information of the torso. Since the spine is the center of the torso in the human body structure, the chest contour points of the torso part of the three-dimensional human body model require the left shoulder key points and the right shoulder key points to divide the three-dimensional human body model into two parts for separate detection.

[0144] In this embodiment, since the chest contour point corresponds to the thorax part of the human body three-dimensional model, and the thorax area is larger, in this step, the first spinal joint point can be the spinal joint point corresponding to the upper end of the thorax or a higher spinal joint point, and the second spinal joint point can be the spinal joint point corresponding to the lower end of the thorax or a lower spinal joint point.

[0145] S62. Establish a horizontal section at the first spinal joint point as a spinal reference section.

[0146] S63. Determine the left shoulder projection point and the right shoulder projection point of the left shoulder joint and the right shoulder joint on the spine reference section according to the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint.

[0147] In this embodiment, in this embodiment, in this embodiment, based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the left shoulder projection point and the right shoulder projection point of the left shoulder joint on the spine reference section are obtained, that is, based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the coordinates of the left shoulder joint in the vertical direction are replaced with the coordinates of the spine reference section in the vertical direction to obtain the coordinates of the left shoulder projection point, and similarly, the coordinates of the right shoulder projection point can be obtained.

[0148] S64. Establish a line connecting the left shoulder projection point and the right shoulder projection point as a reference line.

[0149] S65. Establish a perpendicular line with the first spinal joint point as the orthocenter, and divide the spinal reference section into a first reference section and a second reference section; wherein the perpendicular line is perpendicular to the reference line.

[0150] In this embodiment, since the contour points of the chest cavity may include a left chest cavity and a right chest cavity, and the contour points of the left and right chest cavities may change due to the model established by the user, in this step, the spine reference section is divided into a first reference section and a second reference section according to the left shoulder projection point and the right shoulder projection point.

[0151] S66. Move both the first reference section and the second reference section vertically downward to the second spinal joint point, and obtain the distance between each grid point and the reference line in the first reference section in real time based on the grid information of the trunk, and obtain the distance between each grid point and the reference line in the second reference section in real time.

[0152] In this embodiment, by moving the first reference section and the second reference section vertically downward to the second spinal joint point, the first reference section and the second reference section will successively meet the grid points on the surface of the human body three-dimensional model. In this step, the distance between the grid points appearing on the reference section and the reference line is obtained, that is, the relative position of the grid points on the surface of the human body three-dimensional model can be confirmed.

[0153] S67, taking the grid point with the largest distance from the reference line in the first reference section and the grid point with the largest distance from the reference line in the second reference section as the key points of the torso.

[0154] In this embodiment, the trunk key point finally determined is the largest point of the chest contour in the three-dimensional human body model, that is, the trunk thickness corresponding to the trunk key point is the largest.

[0155] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining the key points of the torso.

[0156] like Figure 8 Specifically, in this embodiment, the method for determining the key points of the torso includes the following steps:

[0157] S71. Determine the three-dimensional information of the hip joint and the three-dimensional information of the first spine joint based on the skeleton information of the trunk.

[0158] In this embodiment, the three-dimensional information of the hip joint in the skeleton information is obtained through the skeleton information of the trunk. In this solution, the hip joint can be the lower end point of the spine, or the connection point between the hip bone and the spine.

[0159] S72. Determine the first spine projection point and the nose tip projection point of the first spine joint point and the nose tip corresponding grid points on the cross section in the horizontal direction of the hip joint point based on the three-dimensional information of the first spine joint point and the grid information of the nose tip corresponding grid points.

[0160] In this embodiment, in this embodiment, based on the three-dimensional information of the first spinal joint point and the grid information of the grid points corresponding to the nose tip, the first spinal projection point and the nose tip projection point of the first spinal joint point and the grid points corresponding to the nose tip are obtained on the cross section in the horizontal direction where the hip joint point is located.

[0161] S73. With the first spine projection point as the origin, three rays are generated on the horizontal cross section where the hip joint point is located; wherein the three rays divide the horizontal cross section where the hip joint point is located into three equal parts, and the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point.

[0162] In this embodiment, taking the first spine projection point as the starting point, three rays of the horizontal cross section where the hip joint is located are generated, the horizontal cross section is divided into three equal parts by the rays, and it is ensured that the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by two of the rays.

[0163] Specifically, find the projection of the first spinal joint point on the horizontal section, find the tip of the nose projection point on the section, connect the first spinal projection point and the tip of the nose projection point, then the direction from the first spinal projection point to the tip of the nose projection point is the reference direction, take the reference direction as 0 degrees, take the first spinal projection point as the origin, emit three rays with angles of 60 degrees, 180 degrees, and 300 degrees respectively, and the above rays can be obtained.

[0164] S74. According to the mesh information of the torso, the mesh point closest to the hip intersection is used as the key point of the torso; wherein the hip intersection is the intersection point of the ray and the mesh structure of the torso.

[0165] In this embodiment, according to the grid information of the trunk, the intersection point of the ray and the grid structure of the trunk is taken as the hip intersection point, and the grid point closest to the hip intersection point is taken as the trunk key point.

[0166] An embodiment of the present invention provides a three-dimensional model construction method. Compared with the construction method provided in the above embodiment, the difference is that this solution provides a method for determining the key points of the torso.

[0167] like Fig. 9 Specifically, in this embodiment, the method for determining the key points of the torso includes the following steps:

[0168] S81. Determine the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint based on the skeleton information of the trunk.

[0169] In this embodiment, the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point are obtained through the skeleton information of the trunk.

[0170] S82. Determine the average of the vertical coordinates of the left hip joint and the right hip joint based on the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint, as the vertical coordinate of the trunk key point.

[0171] In this embodiment, the three-dimensional information of the left hip joint point and the average of the vertical coordinates of the right hip joint point are used as the vertical coordinates of the torso joint point, wherein the torso key points obtained in this embodiment are the torso key points corresponding to the hip contour points.

[0172] S83. Create a horizontal section at the key point of the torso as a reference section for the hip bone.

[0173] S84. The three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point are used to determine the left hip joint point and the left hip joint point and the right hip joint point on the hip reference section.

[0174] In this embodiment, based on the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint, the left hip bone projection point and the right hip bone projection point on the horizontal cross-section where the torso key point is located are obtained. That is, based on the three-dimensional information of the left hip joint, the coordinates of the left hip bone joint in the vertical direction are replaced with the coordinates of the horizontal cross-section in the vertical direction to obtain the coordinates of the left hip bone projection point. Similarly, the coordinates of the right hip bone projection point can be obtained.

[0175] S85. Establish the first ray and the second ray on the hip reference section with the left hip projection point as the starting point; wherein the line connecting the left hip projection point and the right hip bone bisects the angle formed by the first ray and the second ray.

[0176] In this embodiment, the left hip bone projection point is used as the starting point to establish the first ray and the second ray on the hip bone reference section, and ensure that the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the first ray and the second ray.

[0177] S86. Establish a third ray and a fourth ray on the hip reference section with the right hip projection point as the starting point; wherein the line connecting the left hip projection point and the right hip bisects the angle formed by the third ray and the fourth ray.

[0178] In this embodiment, the third ray and the fourth ray are established on the hip reference section with the right hip projection point as the starting point, and it is ensured that the line connecting the left hip projection point and the right hip bisects the angle formed by the third ray and the fourth ray.

[0179] Specifically, for example, taking the projection point of the left hip bone as the origin and the direction from the right hip bone to the left hip bone as 0 degrees, two rays with included angles of 60 degrees and -60 degrees are drawn, which are the first ray and the second ray respectively. Taking the projection point of the right hip bone as the origin and the direction from the left hip bone to the right hip bone as 0 degrees, two rays with included angles of 60 degrees and -60 degrees are drawn, which are the third ray and the fourth ray respectively.

[0180] S87. According to the grid information of the torso, the grid point closest to the intersection point of the hip bones is used as the torso key point; among them, the intersection points of the hip bones are the intersection points of the first ray, the second ray, the third ray and the fourth ray with the grid structure of the torso respectively.

[0181] In this embodiment, according to the grid information of the torso, the intersection point of the ray and the grid structure of the torso is used as the intersection point of the hip bones, and the grid point closest to the intersection point of the hip bones is used as the torso key point.

[0182] The embodiment of the present invention provides a three-dimensional model construction method. In addition to constructing the head proxy and the torso proxy based on the above embodiments, it also includes constructing the proxy of the limbs. Among them, the construction method of the proxy of the limbs includes: obtaining the grid information of the limbs of the human body three-dimensional model; according to the grid information, obtaining the three-dimensional information of the joint points corresponding to the limbs; based on the three-dimensional information, constructing the proxy corresponding to the grid information of the limbs; replacing the limbs in the human body three-dimensional model with the proxy to obtain the simplified and completed human body three-dimensional model.

[0183] Specifically, in this embodiment, the proxy of the limbs can be composed of two hemispheres and a cylinder. Taking the right upper arm between the right shoulder and the right elbow as an example to calculate the generation of the proxy of this part. First, we know the spatial coordinates of the right shoulder and the right elbow, and we use them as the center points c1 and c2 of the two bottom surfaces of the cylinder.

[0184] Next, we need to calculate the radius r c .

[0185] First, we select one of c1 and c2, for example, we select c1. We need to find the reference point set P1, and this set P1 is the vertices of the grids that meet the conditions near c1.

[0186] The vertex p of the grid within the reference point set P1 should satisfy:

[0187] 1. ∠c2c1p < α, where ∠c2c1p is the included angle formed by point c2, point c1, and point p, that is, the included angle formed by point c2, point c1, and point p is less than α, and α is a preset angle.

[0188] 2. d(c1, p) < D, where d(c1, p) is the distance from point c1 to point p, that is, the distance from point c1 to point p is less than D, and D is a preset distance.

[0189] Calculate reference radius l1 is the number of points p in the reference point set P1, d(c1,p) is the distance from point c1 to point p, ∑ pinP1 d(c1,p) is the accumulation of the distances from point c1 to each point p in the reference point set P1.

[0190] Similarly, we can use the above method to obtain a set of reference points P2 that meet the requirements near c2.

[0191] Calculate reference radius l2 is the number of points p in the reference point set P2, d(c2,p) is the distance from point c2 to point p, ∑ pinP2 d(c2,p) is the accumulation of the distances from point c2 to each point p in the reference point set P2.

[0192] Sphere radius r c =w1×r1+w2×r2, w1 and w2 are preset weight values.

[0193] Through the above process, we can calculate the points c1 and c2 of the two bottom centers of the cylinder, and the radius of the bottom of the cylinder is also the radius of the sphere r c , and thus obtain the corresponding proxy body. Similarly, the proxy bodies of the limbs can be obtained. By replacing the limbs of the human body three-dimensional model with the proxy bodies, a simplified human body three-dimensional model can be obtained.

[0194] In a specific embodiment, the present solution provides a method for constructing a three-dimensional model, by constructing proxy bodies of various parts of the three-dimensional human body model, and replacing the corresponding parts of the three-dimensional human body model with the proxy bodies, so as to achieve overall simplification of the three-dimensional human body model. Specifically, it can be achieved by the following method:

[0195] First, the method for constructing proxy bodies of limbs includes: obtaining mesh information of limbs of a three-dimensional human body model; obtaining three-dimensional information of joints corresponding to the limbs based on the mesh information; and constructing proxy bodies corresponding to the mesh information of the limbs based on the three-dimensional information.

[0196] Specifically, in this embodiment, the proxy body of the limbs can be composed of two hemispheres and a cylinder. The right upper arm between the right shoulder and the right elbow is taken as an example to calculate the generation of the partial proxy body. First, we know the spatial coordinates of the right shoulder and the right elbow, and we use them as the two bottom center points c1 and c2 of the cylinder.

[0197] Next we need to calculate the radius r c .

[0198] We select one of c1 and c2, for example, if we select c1, we need to find the reference point set P1, which is the vertices of the mesh near c1 that meet the conditions.

[0199] The vertex p of the grid within the reference point set P1 should satisfy:

[0200] 1. ∠c2c1p < α, where ∠c2c1p is the included angle formed by points c2, c1, and p, that is, the included angle formed by points c2, c1, and p is less than α, and α is a preset angle.

[0201] 2. d(c1, p) < D, where d(c1, p) is the distance from point c1 to point p, that is, the distance from point c1 to point p is less than D, and D is a preset distance.

[0202] Calculate the reference radius l1 is the number of points p in the reference point set P1, d(c1, p) is the distance from point c1 to point p, ∑ pinP1 d(c1, p) is the sum of the distances from point c1 to each point p in the reference point set P1.

[0203] Similarly, we can obtain the reference point set P2 that meets the conditions near c2 through the above method.

[0204] Calculate the reference radius l2 is the number of points p in the reference point set P2, d(c2, p) is the distance from point c2 to point p, ∑ pinP2 d(c2, p) is the sum of the distances from point c2 to each point p in the reference point set P2.

[0205] Sphere radius r c = w1 × r1 + w2 × r2, where w1 and w2 are preset weight values.

[0206] Through the above process, the calculation of the points c1 and c2, which are the centers of the two bottom surfaces of the cylinder, and the radius of the bottom surface of the cylinder, which is also the sphere radius r, can be completed. c , and the corresponding proxy body can be obtained therefrom. Similarly, the proxy bodies of the limbs can be obtained.

[0207] Secondly, the method for constructing the proxy bodies of the torso and the head includes: obtaining the mesh information and bone information of the head and torso of the human body three-dimensional model; determining the key points of the head and the key points of the torso according to the mesh information and the bone information; constructing the head proxy body according to the key points of the head, and constructing the torso proxy body according to the key points of the torso.

[0208] Specifically, in this embodiment, regarding the proxy body of the head, we reconstruct the mesh structure and select the key points of the head (the original mesh vertices closest to the key points), specifically, for example, eyes (two points), nose tip (one point), mouth corners (two points), chin tip (one point), ears (two points), top of the head (one point), head circumference (two points). The above key points are only a specific implementation method of this solution, and other methods can also be used to select key points to achieve simplification of the head.

[0209] Specifically, in this embodiment, regarding the proxy body of the torso, we reconstruct the mesh structure and select the key points of the torso (the original mesh vertices closest to the key points), specifically, for example, the torso key points located in the shoulder area, the torso key points located in the chest, the torso key points located in the hips, and the torso key points located in the hips. The above key points are only a specific implementation method of this solution, and other methods can also be used to select key points to achieve the corresponding torso simplification.

[0210] Among them, the eyes, nose tip, mouth corners, chin tip, ears, and top of the head can be directly obtained in the human body 3D model.

[0211] The key points corresponding to the head circumference can be determined by the following method: according to the bone information of the head, the three-dimensional information of the neck joint points is obtained; according to the average of the coordinates of the grid points corresponding to the top of the head and the tip of the nose in the vertical direction, the coordinates of the head circumference in the vertical direction are taken; according to the three-dimensional information of the neck joint points and the grid information of the grid points corresponding to the tip of the nose, the neck projection points and the nose tip projection points of the grid points corresponding to the neck joint points and the nose tip on the cross section in the horizontal direction of the head circumference are determined; with the neck projection point as the origin, four rays are generated on the cross section in the horizontal direction of the head circumference; wherein, the four rays divide the cross section in the horizontal direction of the head circumference into four equal parts, and the line connecting the neck projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; according to the grid information of the head, the grid point closest to the head circumference intersection is taken as the grid point corresponding to the head circumference; wherein, the head circumference intersection is the intersection of the ray and the grid structure of the head.

[0212] The key points corresponding to the back of the head can be determined by the following method: based on the bone information of the head, the three-dimensional information of the neck joint points is obtained; based on the three-dimensional information of the neck joint points, the neck projection point of the neck joint points on the horizontal cross-section where the grid point corresponding to the nose tip is located is determined; based on the grid information of the head, the grid point closest to the intersection of the back of the head is used as the grid point corresponding to the back of the head; wherein the intersection of the back of the head is the intersection of the extended line of the line connecting the grid point corresponding to the nose tip and the neck projection point and the grid structure of the head.

[0213] The key points of the trunk located in the shoulder area can be determined by the following method: based on the skeleton information of the trunk, the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint and the three-dimensional information of the second spine joint are determined; based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the average of the coordinates of the left shoulder joint and the right shoulder joint in the vertical direction is determined as the coordinate of the key points of the trunk in the vertical direction; based on the three-dimensional information of the second spine joint and the grid information of the grid points corresponding to the nose tip, the second spine projection point and the nose tip projection point of the second spine joint and the grid points corresponding to the nose tip on the cross section in the horizontal direction where the key points of the trunk are located are determined; based on the three-dimensional information of the left shoulder joint The left shoulder projection point and the right shoulder projection point on the horizontal cross section of the trunk key point are determined based on the three-dimensional information of the left shoulder joint and the right shoulder joint; with the second spine projection point as the starting point, according to the second spine projection point and the nose tip projection point, a first ray is established from the second spine projection point to the nose tip projection point; according to the grid information of the trunk, the grid point closest to the trunk intersection is used as the trunk key point; wherein, the trunk intersection is the intersection of the second ray constructed with the left shoulder projection point as the starting point and the grid structure of the trunk, and the trunk intersection is the intersection of the third ray constructed with the right shoulder projection point as the starting point and the grid structure of the trunk, and the directions of the first ray, the second ray and the third ray are the same.

[0214] The key points of the trunk located at the chest can be determined by the following method: based on the skeletal information of the trunk, the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spine joint and the three-dimensional information of the second spine joint are determined; wherein the coordinate of the first spine joint in the vertical direction is greater than the coordinate of the second spine joint in the vertical direction; a cross section in the horizontal direction of the first spine joint is established as a spine reference cross section; based on the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint, the left shoulder projection point and the right shoulder projection point of the left shoulder joint and the right shoulder joint on the spine reference cross section are determined; a connection is established between the left shoulder projection point and The line connecting the right shoulder projection points is used as the reference line; a vertical line is established with the first spinal joint point as the orthocenter to divide the spinal reference section into a first reference section and a second reference section; wherein the vertical line is perpendicular to the reference line; both the first reference section and the second reference section are moved vertically downward to the second spinal joint point, and according to the grid information of the trunk, the distance between each grid point in the first reference section and the reference line is obtained in real time, and the distance between each grid point in the second reference section and the reference line is obtained in real time; the grid point with the largest distance from the reference line in the first reference section and the grid point with the largest distance from the reference line in the second reference section are used as the key points of the trunk.

[0215] The key point of the trunk located at the hip can be determined by the following method: based on the skeleton information of the trunk, determine the three-dimensional information of the hip joint point and the three-dimensional information of the first spine joint point; based on the three-dimensional information of the first spine joint point and the grid information of the grid point corresponding to the nose tip, determine the first spine projection point and the nose tip projection point of the first spine joint point and the grid point corresponding to the nose tip on the horizontal cross-section of the hip joint point; with the first spine projection point as the origin, generate three rays on the horizontal cross-section of the hip joint point; wherein the three rays divide the horizontal cross-section of the hip joint point into three equal parts, and the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; based on the grid information of the trunk, take the grid point closest to the hip intersection as the key point of the trunk; wherein the hip intersection is the intersection of the ray and the grid structure of the trunk.

[0216] The key points of the trunk located at the hip can be determined by the following method: based on the skeleton information of the trunk, determine the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point; based on the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point, determine the average of the vertical coordinates of the left hip joint point and the right hip joint point as the vertical coordinates of the key points of the trunk; establish a horizontal section at the key points of the trunk as a hip reference section; based on the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point, determine the left hip projection point and the left hip joint point on the hip reference section. Right hip bone projection point; establish the first ray and the second ray on the hip bone reference section with the left hip bone projection point as the starting point; wherein, the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the first ray and the second ray; establish the third ray and the fourth ray on the hip bone reference section with the right hip bone projection point as the starting point; wherein, the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the third ray and the fourth ray; according to the grid information of the torso, the grid point closest to the hip bone intersection is used as the key point of the torso; wherein, the hip bone intersection is the intersection of the first ray, the second ray, the third ray and the fourth ray with the grid structure of the torso respectively.

[0217] Through the above-mentioned head key points and torso key points, the corresponding head proxy body and torso proxy body are constructed respectively.

[0218] In this embodiment, constructing a proxy body through head key points and torso key points can be achieved in the following manner: connecting the head key points in pairs to obtain a first grid structure; removing the lines inside the first grid structure to obtain a head grid structure as a head proxy body; connecting the torso key points in pairs to obtain a second grid structure; removing the lines inside the second grid structure to obtain a torso grid structure as a torso proxy body.

[0219] Finally, the proxy bodies of the limbs, the head and the torso constructed in the above embodiments are used to replace various parts of the three-dimensional human body model, thereby simplifying the three-dimensional human body model. By performing collision detection on the simplified three-dimensional human body model, proxy calculation of the three-dimensional human body model can be quickly realized, thereby improving data processing efficiency.

[0220] like Fig.10 As shown, an embodiment of the present invention provides a three-dimensional model construction device, which includes: a first processing unit, a second processing unit, a third processing unit and a fourth processing unit.

[0221] In this embodiment, the first processing unit is used to obtain mesh information and bone information of the head and torso of the three-dimensional human body model.

[0222] In this embodiment, the second processing unit is used to determine the key points of the head and the key points of the torso according to the mesh information and the skeleton information.

[0223] In this embodiment, the third processing unit is used to construct a head proxy body according to the head key points, and to construct a torso proxy body according to the torso key points.

[0224] In this embodiment, the fourth processing unit is used to replace the head and torso of the three-dimensional human body model with the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

[0225] In this embodiment, the key points of the head include: mesh points corresponding to the eyes, nose tip, mouth corners, chin tip, ears, top of the head, head circumference and back of the head, respectively.

[0226] In this embodiment, the second processing unit is specifically used to obtain three-dimensional information of the neck joint point according to the bone information of the head; use the average of the coordinates of the grid points corresponding to the top of the head and the tip of the nose in the vertical direction as the coordinates of the head circumference in the vertical direction; determine the neck projection point and the nose tip projection point of the grid points corresponding to the neck joint point and the nose tip on the cross section in the horizontal direction of the head circumference according to the three-dimensional information of the neck joint point and the grid information of the grid points corresponding to the nose tip; take the neck projection point as the origin, and generate four rays on the cross section in the horizontal direction of the head circumference; wherein the four rays divide the cross section in the horizontal direction of the head circumference into four equal parts, and the line connecting the neck projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; according to the grid information of the head, use the grid point closest to the head circumference intersection as the grid point corresponding to the head circumference; wherein the head circumference intersection is the intersection of the ray and the grid structure of the head.

[0227] In this embodiment, the second processing unit is specifically used to obtain three-dimensional information of neck joints based on the bone information of the head; determine the neck projection point of the neck joints on the horizontal cross-section where the grid point corresponding to the nose tip is located based on the three-dimensional information of the neck joints; based on the grid information of the head, use the grid point closest to the intersection with the back of the head as the grid point corresponding to the back of the head; wherein the intersection with the back of the head is the intersection of the extended line of the line connecting the grid point corresponding to the nose tip and the neck projection point and the grid structure of the head.

[0228] In this embodiment, the second processing unit is specifically used to determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint and the three-dimensional information of the second spinal joint according to the skeletal information of the trunk; determine the mean of the vertical coordinates of the left shoulder joint and the right shoulder joint according to the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint as the coordinate of the trunk key point in the vertical direction; determine the second spinal projection point and the nose tip projection point of the second spinal joint and the grid point corresponding to the nose tip on the cross section in the horizontal direction of the trunk key point according to the three-dimensional information of the second spinal joint and the grid information of the grid point corresponding to the nose tip ... The three-dimensional information of the right shoulder joint is used to determine the left shoulder projection point and the right shoulder projection point on the cross section in the horizontal direction of the trunk key point. Taking the second spine projection point as the starting point, according to the second spine projection point and the nose tip projection point, a first ray is established from the second spine projection point to the nose tip projection point. According to the grid information of the trunk, the grid point closest to the trunk intersection is used as the trunk key point. Among them, the trunk intersection is the intersection of the second ray constructed with the left shoulder projection point as the starting point and the grid structure of the trunk, and the trunk intersection is the intersection of the third ray constructed with the right shoulder projection point as the starting point and the grid structure of the trunk, and the directions of the first ray, the second ray and the third ray are the same.

[0229] In this embodiment, the second processing unit is specifically used to determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spine joint and the three-dimensional information of the second spine joint according to the skeletal information of the trunk; wherein the coordinates of the first spine joint in the vertical direction are greater than the coordinates of the second spine joint in the vertical direction; establish a cross section in the horizontal direction of the first spine joint as a spine reference cross section; determine the left shoulder projection point and the right shoulder projection point of the left shoulder joint and the right shoulder joint on the spine reference cross section according to the three-dimensional information of the left shoulder joint and the three-dimensional information of the right shoulder joint; establish a connection between the left shoulder projection point and the right The line connecting the shoulder projection points is used as the reference line; a vertical line is established with the first spinal joint point as the orthocenter to divide the spinal reference section into a first reference section and a second reference section; wherein the vertical line is perpendicular to the reference line; the first reference section and the second reference section are both moved vertically downward to the second spinal joint point, and according to the grid information of the trunk, the distance between each grid point in the first reference section and the reference line is obtained in real time, and the distance between each grid point in the second reference section and the reference line is obtained in real time; the grid point with the largest distance from the reference line in the first reference section and the grid point with the largest distance from the reference line in the second reference section are used as the key points of the trunk.

[0230] In this embodiment, the second processing unit is specifically used to determine the three-dimensional information of the hip joint and the three-dimensional information of the first spine joint according to the skeleton information of the trunk; determine the first spine projection point and the nose tip projection point of the first spine joint and the grid points corresponding to the nose tip on the horizontal cross-section of the hip joint according to the three-dimensional information of the first spine joint and the grid information of the grid points corresponding to the nose tip; with the first spine projection point as the origin, three rays are generated on the horizontal cross-section of the hip joint; wherein the three rays divide the horizontal cross-section of the hip joint into three equal parts, and the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; according to the grid information of the trunk, the grid point closest to the hip intersection is used as the trunk key point; wherein the hip intersection is the intersection of the ray and the grid structure of the trunk.

[0231] In this embodiment, the second processing unit is specifically used to determine the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint according to the skeletal information of the trunk; determine the mean of the vertical coordinates of the left hip joint and the right hip joint according to the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint as the coordinates of the trunk key points in the vertical direction; establish a cross section in the horizontal direction at the trunk key points as a hip reference cross section; determine the left hip projection point and the right hip projection point of the left hip joint and the left hip joint on the hip reference cross section based on the three-dimensional information of the left hip joint and the three-dimensional information of the right hip joint. bone projection point; taking the left hip bone projection point as the starting point, establishing the first ray and the second ray on the hip bone reference section; wherein, the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the first ray and the second ray; taking the right hip bone projection point as the starting point, establishing the third ray and the fourth ray on the hip bone reference section; wherein, the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the third ray and the fourth ray; according to the grid information of the torso, the grid point closest to the hip bone intersection is used as the key point of the torso; wherein, the hip bone intersection is the intersection of the first ray, the second ray, the third ray and the fourth ray with the grid structure of the torso, respectively.

[0232] In this embodiment, the third processing unit is specifically used to connect the head key points in pairs to obtain a first grid structure; remove the connecting lines inside the first grid structure to obtain a head grid structure as a head proxy body; connect the torso key points in pairs to obtain a second grid structure; remove the connecting lines inside the second grid structure to obtain a torso grid structure as a torso proxy body.

[0233] like Fig.11 As shown, an embodiment of the present invention provides a three-dimensional model construction system, including a processor 1110, a communication interface 1120, a memory 1130 and a communication bus 1140, wherein the processor 1110, the communication interface 1120, and the memory 1130 communicate with each other through the communication bus 1140;

[0234] Memory 1130, used for storing computer programs;

[0235] The processor 1110 is used to implement the following steps when executing the program stored in the memory 1130:

[0236] Obtain mesh information and bone information of the head and torso of the human body 3D model;

[0237] Determine the key points of the head and torso based on the mesh information and skeleton information;

[0238] According to the key points of the head, a head proxy body is constructed, and according to the key points of the torso, a torso proxy body is constructed;

[0239] The head and torso of the human body 3D model are replaced by the head proxy body and the torso proxy body to obtain a simplified and completed human body 3D model.

[0240] In the electronic device provided by the embodiment of the present invention, the processor 1110 executes the program stored in the memory 1130 to obtain the mesh information and skeleton information of the head and torso of the human body three-dimensional model, and determines the key points of the head and torso according to the mesh information and the skeleton information, and constructs a proxy body corresponding to the head and torso according to the key points of the head and the torso, and replaces the corresponding parts in the human body three-dimensional model with the proxy body to simplify the human body three-dimensional model, and completes the motion redirection, collision calculation, penetration repair and other algorithms of low-precision scenes through the simplified human body three-dimensional model, avoiding the above-mentioned calculation detection with the original human body three-dimensional model, thereby improving the calculation efficiency.

[0241] The communication bus 1140 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0242] The communication interface 1120 is used for communication between the above electronic device and other devices.

[0243] The memory 1130 may include a random access memory 1130 (RAM), or may include a non-volatile memory 1130 (non-volatile memory), such as at least one disk memory 1130. Optionally, the memory 1130 may also be at least one storage device located away from the processor 1110.

[0244] The above-mentioned processor 1110 can be a general-purpose processor 1110, including a central processing unit 1110 (CPU), a network processor 1110 (NP), etc.; it can also be a digital signal processor 1110 (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0245] An embodiment of the present invention provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors 1110 to implement the three-dimensional model construction method of any of the above embodiments.

[0246] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk SolidStateDisk (SSD)), etc.

[0247] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional model construction method, characterized in that: The construction method comprises: Obtain mesh information and bone information of the head and torso of the human body 3D model; Determine the key points of the head and the key points of the torso according to the mesh information and the skeleton information; A head proxy body is constructed according to the head key points, and a torso proxy body is constructed according to the torso key points, including: connecting the head key points in pairs to obtain a first grid structure; removing the lines inside the first grid structure to obtain a head grid structure as the head proxy body; connecting the torso key points in pairs to obtain a second grid structure; removing the lines inside the second grid structure to obtain a torso grid structure as the torso proxy body; The head and torso of the three-dimensional human body model are replaced by the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

2. The construction method according to claim 1, characterized in that: The key points of the head include: mesh points corresponding to the eyes, nose tip, mouth corners, chin tip, ears, top of the head, head circumference and back of the head respectively.

3. The construction method according to claim 2, characterized in that: Methods for determining the grid points corresponding to the head circumference include: Acquire three-dimensional information of neck joints according to the bone information of the head; The average of the vertical coordinates of the grid points corresponding to the top of the head and the tip of the nose is used as the vertical coordinate of the head circumference; Determine, according to the three-dimensional information of the neck joint point and the grid information of the grid point corresponding to the nose tip, the neck projection point and the nose tip projection point of the neck joint point and the grid point corresponding to the nose tip on the cross section in the horizontal direction of the head circumference; Taking the neck projection point as the origin, four rays are generated on the cross section in the horizontal direction where the head circumference is located; wherein the four rays divide the cross section in the horizontal direction where the head circumference is located into four equal parts, and the line connecting the neck projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; According to the grid information of the head, the grid point closest to the head circumference intersection is used as the grid point corresponding to the head circumference; wherein the head circumference intersection is the intersection of the ray and the grid structure of the head.

4. The construction method according to claim 2, characterized in that: Methods for determining the grid points corresponding to the back of the head include: Acquire three-dimensional information of neck joints according to the bone information of the head; Determine, according to the three-dimensional information of the neck joint point, a neck projection point of the neck joint point on a cross section in a horizontal direction where a grid point corresponding to the nose tip is located; According to the grid information of the head, the grid point closest to the intersection of the back of the head is used as the grid point corresponding to the back of the head; wherein the intersection of the back of the head is the intersection of the extended line of the line connecting the grid point corresponding to the nose tip and the neck projection point and the grid structure of the head.

5. The construction method according to claim 2, characterized in that: Methods for determining torso key points include: Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, and the three-dimensional information of the second spine joint according to the skeleton information of the trunk; According to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point, determine the average of the coordinates of the left shoulder joint point and the right shoulder joint point in the vertical direction as the coordinate of the trunk key point in the vertical direction; Determine, based on the three-dimensional information of the second spinal joint point and the grid information of the grid point corresponding to the nose tip, the second spinal projection point and the nose tip projection point of the second spinal joint point and the grid point corresponding to the nose tip on the cross section in the horizontal direction where the trunk key point is located; Determine, according to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point, a left shoulder projection point and a right shoulder projection point of the left shoulder joint point and the right shoulder joint point on a cross section in a horizontal direction where the key point of the trunk is located; Taking the second spine projection point as the starting point, and according to the second spine projection point and the nose tip projection point, establishing a first ray from the second spine projection point to the nose tip projection point; According to the mesh information of the torso, the mesh point closest to the torso intersection is used as the torso key point; wherein the torso intersection is the intersection of a second ray constructed with the left shoulder projection point as the starting point and the mesh structure of the torso, and the torso intersection is the intersection of a third ray constructed with the right shoulder projection point as the starting point and the mesh structure of the torso, and the directions of the first ray, the second ray and the third ray are the same.

6. The construction method according to claim 2, characterized in that: Methods for determining torso key points include: Determine the three-dimensional information of the left shoulder joint, the three-dimensional information of the right shoulder joint, the three-dimensional information of the first spinal joint, and the three-dimensional information of the second spinal joint according to the skeleton information of the trunk; wherein the coordinate of the first spinal joint in the vertical direction is greater than the coordinate of the second spinal joint in the vertical direction; Establishing a cross section in the horizontal direction of the first spinal joint point as a spinal reference cross section; Determine a left shoulder projection point and a right shoulder projection point of the left shoulder joint point and the right shoulder joint point on the spine reference section according to the three-dimensional information of the left shoulder joint point and the three-dimensional information of the right shoulder joint point; Establishing a line connecting the left shoulder projection point and the right shoulder projection point as a reference line; Establishing a vertical line with the first spinal joint point as the orthocenter to divide the spinal reference section into a first reference section and a second reference section; wherein the vertical line is perpendicular to the reference line; Move both the first reference section and the second reference section vertically downward to the second spinal joint point, and obtain the distance between each grid point in the first reference section and the reference line in real time according to the grid information of the trunk, and obtain the distance between each grid point in the second reference section and the reference line in real time; The grid point with the largest distance from the reference line in the first reference section and the grid point with the largest distance from the reference line in the second reference section are used as the key points of the torso.

7. The construction method according to claim 2, characterized in that: Methods for determining torso key points include: Determine the three-dimensional information of the hip joint and the three-dimensional information of the first spine joint according to the skeleton information of the trunk; Determine, based on the three-dimensional information of the first spinal joint point and the grid information of the grid points corresponding to the nose tip, a first spinal projection point and a nose tip projection point of the grid points corresponding to the first spinal joint point and the nose tip on a cross section in the horizontal direction where the hip joint point is located; Taking the first spine projection point as the origin, three rays are generated on the cross section in the horizontal direction where the hip joint point is located; wherein the three rays divide the cross section in the horizontal direction where the hip joint point is located into three equal parts, and the line connecting the first spine projection point and the nose tip projection point bisects the angle formed by the two rays adjacent to the nose tip projection point; According to the mesh information of the torso, the mesh point closest to the hip intersection is used as the key point of the torso; wherein the hip intersection is the intersection point of the ray and the mesh structure of the torso.

8. The construction method according to claim 2, characterized in that: Methods for determining torso key points include: Determine the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point according to the skeleton information of the torso; According to the three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point, determine the average of the coordinates of the left hip joint point and the right hip joint point in the vertical direction as the coordinate of the trunk key point in the vertical direction; Establishing a cross section in the horizontal direction at the key point of the torso as a reference cross section for the hip bone; The three-dimensional information of the left hip joint point and the three-dimensional information of the right hip joint point are used to determine the left hip joint point and the left hip joint point and the right hip joint point on the hip reference section; A first ray and a second ray located on the hip bone reference section are established with the left hip bone projection point as a starting point; wherein a line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the first ray and the second ray; A third ray and a fourth ray located on the hip bone reference section are established with the right hip bone projection point as the starting point; wherein the line connecting the left hip bone projection point and the right hip bone bisects the angle formed by the third ray and the fourth ray; According to the grid information of the torso, the grid point closest to the hip intersection is used as the key point of the torso; wherein the hip intersection is the intersection of the first ray, the second ray, the third ray and the fourth ray with the grid structure of the torso respectively.

9. A three-dimensional model building device, characterized in that: The construction device comprises: A first processing unit is used to obtain mesh information and bone information of a head and a torso of a three-dimensional human body model; A second processing unit, configured to determine a head key point and a torso key point according to the mesh information and the skeleton information; The third processing unit is used to construct a head proxy body according to the head key points, and to construct a torso proxy body according to the torso key points, including: connecting the head key points in pairs to obtain a first grid structure; removing the lines inside the first grid structure to obtain a head grid structure as the head proxy body; connecting the torso key points in pairs to obtain a second grid structure; removing the lines inside the second grid structure to obtain a torso grid structure as the torso proxy body; The fourth processing unit is used to replace the head and the torso of the three-dimensional human body model with the head proxy body and the torso proxy body to obtain a simplified three-dimensional human body model.

10. A three-dimensional model construction system, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the three-dimensional model construction method described in any one of claims 1 to 8 when executing the program stored in the memory.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the three-dimensional model construction method described in any one of claims 1 to 8.

Images