A digital human muscle model pose generation method and device

Abstract

The application relates to the technical field of digital human body models, in particular to a digital human body muscle model posture generation method and device. The method comprises the following steps: obtaining a tomographic image of an initial posture of a human body by using high-precision CT, and inversely modeling a geometric model of human body bones and muscles according to the tomographic image; calculating a first motion influence weight of each muscle node attached to a bone node according to each muscle node and the bone node; calculating a second motion influence weight of each muscle node according to the influence of adjacent muscle nodes on each muscle node; calculating a comprehensive weight of each muscle node according to the first motion influence weight and the second motion influence weight; and calculating a Laplace coordinate of a next moment of a current muscle node according to the comprehensive weight of the current muscle node, the position of the current muscle node and adjacent muscle nodes. The application generates a high-fidelity muscle geometric model by using a Laplace mesh deformation constraint technology.

Description

Technical Field

[0001] This application relates to the field of digital human body model technology, and more specifically, to a method and device for generating poses of a digital human muscle model. Background Technology

[0002] High-precision digital human models are crucial in fields such as automotive safety crash simulation. Existing methods for generating models of different postures have the following limitations:

[0003] 1. Limitations of Skeletal Rigid Transformation: While hierarchical coordinate system-based skeletal rigid transformation can efficiently determine the pose frame, it is only applicable to skeletons. Directly applying it to soft tissues such as muscles leads to unnatural deformation and volume loss, failing to meet the high requirements of biomechanical simulation for the realism of soft tissue morphology.

[0004] 2. Defects of traditional skinning techniques: Traditional methods such as linear blending skinning are prone to distortions such as "candy wrapping" at joints. Although circle fitting smoothing algorithms can smooth the mesh, they are designed to solve the "jaw" problem of the mesh, rather than to handle large-scale pose deformations driven by the underlying skeleton. Furthermore, traditional circle fitting algorithms suffer from mesh shrinkage during the smoothing process.

[0005] 3. Limitations of Non-rigid Registration Techniques: Thin-plate spline algorithms are powerful tools for handling non-rigid point set registration. However, these methods are typically used for global registration of images or point sets, and their energy function minimization process lacks fine-grained control for muscle deformations that require precise adherence to anatomical attachment relationships (i.e., bone-driven).

[0006] 4. Statistical shape models can effectively capture anatomical shape variations in populations. However, they primarily describe shape differences between individuals, rather than nonlinear soft tissue deformation of the same individual in different postures.

[0007] Therefore, there is an urgent need in the field for an automated method that can seamlessly integrate precise bone actuation with biomechanically sound, locally detailed soft tissue deformation. Summary of the Invention

[0008] The purpose of this application is to provide a method and device for generating digital human muscle model poses. By using Laplace mesh deformation constraint technology, a high-fidelity geometric model of muscles can be generated, which is particularly suitable for high-quality construction scenarios of human geometric models and finite element models in automotive safety and ergonomics.

[0009] To achieve the above objectives, this application adopts the following technical solution:

[0010] Firstly, this application provides a method for generating poses of a digital human muscle model, including:

[0011] High-precision CT scans are used to acquire tomographic images of the human body in its initial posture, and geometric models of the human skeleton and muscles are obtained by reverse modeling based on the tomographic images; the geometric model includes skeletal nodes and muscle nodes, as well as the connection relationships between the nodes;

[0012] Based on each muscle node and the bone node to which it is attached, calculate the weight of the first motion influence of the attached bone on each muscle node.

[0013] Calculate the second motion influence weight of each muscle node based on the influence of adjacent muscle nodes on each muscle node;

[0014] Calculate the comprehensive weight of each muscle node based on the first motion influence weight and the second motion influence weight;

[0015] Based on the overall weight of the current muscle node and the positions of the current muscle node and its adjacent muscle nodes, calculate the Laplace coordinates of the current muscle node at the next moment.

[0016] Secondly, this application provides an electronic device, comprising:

[0017] At least one processor, and a memory communicatively connected to at least one of the processors;

[0018] The memory stores instructions that can be executed by at least one of the processors, which are executed by at least one of the processors to enable the at least one of the processors to perform any of the digital human muscle model pose generation methods.

[0019] The embodiments of this application have the following technical effects:

[0020] 1. High fidelity: By constraining the Laplace coordinates and the shape of the muscle cross section, the generated muscle shape is natural and the local details are rich, effectively avoiding the distortion and volume loss problems of traditional skinning technology.

[0021] 2. High efficiency and automation: It enables automatic muscle generation from end to end, minimizing the workload of manual adjustments, greatly improving the development efficiency of digital human body models and shortening the development cycle.

[0022] 3. Anatomical rationality: Weight mapping and energy constraints based on anatomical knowledge make muscle deformation (such as stretching, contraction, and bulging) more consistent with real biomechanical behavior.

[0023] 4. Technological Integration and Innovation: By deeply integrating Laplace deformation from computer graphics with biomechanics and anatomy, a novel and reliable solution is provided for the specific problem of generating poses in digital human models. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating a method for generating poses of a digital human muscle model provided in an embodiment of this application.

[0026] Figure 2 This is a schematic diagram of the relative positions of muscles and bones provided in the embodiments of this application;

[0027] Figure 3 This is a schematic diagram of the approximate function provided in the embodiments of this application;

[0028] Figure 4 This is a schematic diagram of the current muscle node v0 and adjacent muscle nodes v1~v5 provided in the embodiments of this application;

[0029] Figure 5 This is a schematic diagram illustrating the changes in the position of muscle nodes provided in an embodiment of this application;

[0030] Figure 6 This is a schematic diagram of the unnatural deformation of a muscle node provided in an embodiment of this application;

[0031] Figure 7 This is a schematic diagram of the deformation of the gluteus maximus provided in the embodiments of this application;

[0032] Figure 8 These are comparison images of the smoothed protruding muscle nodes provided in the embodiments of this application;

[0033] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0034] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0035] Figure 1This is a flowchart illustrating a method for generating poses of a digital human muscle model provided in this application embodiment. This embodiment is applicable to scenarios involving digital modeling of human bones and muscles. Figure 1 The method shown is performed by an electronic device.

[0036] See Figure 1 The methods provided in this application include:

[0037] S110. High-precision CT is used to obtain tomographic images of the human body in its initial posture, and geometric models of the human skeleton and muscles are obtained by reverse modeling based on the tomographic images; the geometric model includes bone nodes and muscle nodes, as well as the connection relationships between the nodes.

[0038] High-precision CT (Computed Tomography) scans were used to obtain tomographic images of the human body in its initial posture (e.g., standard anatomical standing posture, neutral position), covering the major skeletal and muscular regions of the entire body. Based on the grayscale differences in the tomographic images, a threshold segmentation method was used to extract the boundaries of tissues such as bones and muscles, and a digital human skeletal and muscular geometric model was generated through reverse modeling. A triangular mesh model was created on the surface of the geometric model. The skeleton is a rigid triangular mesh model, and the muscles are deformable triangular mesh models. Here, "deformable" means that the side lengths and the included angles between adjacent sides can be changed. The specific process is as follows:

[0039] The first step involves scanning the human body using a Micro-CT scanner to obtain CT tomographic images with the required clarity. Different gray levels in the tomographic images represent the boundaries of different tissues in the human body. 3D image processing software can then be used to obtain 3D models of different tissues. The second step involves using 3D image processing software to perform gray-scale segmentation on the tomographic images. The segmentation accuracy depends on the scanning resolution. CT images are represented as cubic units of different gray levels in the 3D image processing software. Selecting and extracting cubic units of the same gray level yields a 3D model of a single tissue or organ. However, when only a portion of the tissue or organ's boundary falls within a cubic unit, errors, known as volume effects, are unavoidable. In this case, manual preprocessing is required to check and repair the surface smoothness of the generated 3D model. The third step involves the 3D image processing software outputting the segmented model as an STL format and importing it into a geometric surface reconstruction software. This converts the 3D model into a NURBS surface (Non-Uniform Rational B-Spline Surface), which has the advantage of converting the surface features of a polygonal model into a finite number of triangular patches. The final geometric model's outer contour surface consists of a series of triangular facets. The number and size of these facets depend on the required accuracy of the model. The data formed by each vertex of a facet is the point cloud data, recording the spatial coordinates of all points. The model is then further refined using the NURBS surface reconstruction software to remove or reassign unsuitable facets before outputting the final result.

[0040] After obtaining the geometric model, a list of bone and muscle associations is generated. This list includes the bone nodes to which each muscle node is attached, as well as the absolute coordinates of each bone node. Based on the structure of human muscles, strip-shaped muscles are attached to bones at both ends; therefore, all muscle nodes on a single muscle are attached to the same set of bone nodes. The absolute coordinates of each bone node are relative to the body's center of mass. After obtaining the geometric model, the relative position of each muscle node with respect to its attached (any) bone node can be determined. Combined with the absolute coordinates of the bone nodes, the absolute coordinates of each muscle node can be obtained.

[0041] S120. Based on each muscle node and the bone node to which it is attached, calculate the weight of the first motion influence of the attached bone on each muscle node.

[0042] For each muscle node, the weights of the influence of the attached skeletal nodes are calculated. These weights are used to quantify the driving force of the movement of the skeletal nodes to which the muscle is attached, providing a skeletal-muscle mechanical transmission relationship for subsequent deformation.

[0043] In this embodiment, muscles are considered to be strip-shaped, and the part of the muscle connected to the bone is associated with the part of the bone that is in contact with it. Then, the entire muscle is considered to be associated with the bone to which it is attached.

[0044] At both ends of a muscle, it connects to the bone. Considering the minimal relative movement between the muscle and bone at the attachment / connection points, this can be approximated as a rigid connection, and thus carries a higher weight. However, near the middle of the muscle, further from the attachment / connection points, the influence of the relative position of the bone on the muscle is significantly less than at the ends. Figure 2 As shown, the muscle nodes at the upper and lower ends ( It can be considered that it moves together with the skeleton, and the influence weight of the first movement is 1. This is the straight-line distance from the muscle node to the endpoint of the muscle-bone connection. Figure 2 In the middle, the endpoint is the end where the upper arm bone connects to the shoulder. This represents the total length of the muscle.

[0045] In one specific implementation, for a muscle node attached to a skeletal node, a first motion influence weight is set to be greater than a set value, which may be 0.9 and the first motion influence weight may be 1.

[0046] In another specific implementation, for muscle nodes not attached to skeletal nodes, a first motion influence weight is determined based on the weight distribution characteristics of the muscle's spindle-shaped structure. Specifically, the free zone in the middle of the muscle is less driven by skeletal movement and conforms to the weight distribution characteristics of the muscle's spindle-shaped structure. In this case, the first motion influence weight can be a value less than 1, as shown in the following formula:

[0047] ;

[0048] The straight-line distance from the endpoint where the muscle connects to the bone is... The first movement at the muscle node affects the weight. This is a function related to muscle weight distribution. Let be the straight-line distance from the muscle node to the endpoint of the muscle-bone connection. This represents the total length of the muscle.

[0049] Because muscles are spindle-shaped, their cross-sectional area varies gradually along the direction of force application, resulting in a variation in muscle weight across each cross-section. To describe the effect of this phenomenon on weights using an approximate function, an accurate description of the muscle shape is needed:

[0050] ;

[0051] for The radius of the muscle cross-section is approximately a circle.

[0052] Optionally, the function curve is as follows: Figure 3 As shown, the horizontal axis is The function is expressed as a multiple of the total muscle length L, and the total coordinate is the radius y of the muscle interface. It's easy to see that this function has its inverse. (about The figure enclosed by symmetry, that is... Figure 3 The shaded area in the graph effectively depicts the spindle-shaped structure of muscles. Since the cross-sectional weight is positively correlated with the square of the cross-sectional radius, the correlation function for muscle weight distribution is... It can be represented as:

[0053] ;

[0054] in, This is a function related to muscle weight distribution. The coefficient related to muscle type and condition is a value no less than 1. This is because different muscle parts have different water and protein contents, which inevitably affects the cross-sectional density. A certain part of the muscle has a relatively higher density, so the density at a certain cross-section of that muscle will also be relatively higher, making... Larger is considered larger; conversely, smaller is considered smaller. Furthermore, the muscle's state (such as contraction or relaxation) also affects its size. The value of .

[0055] S130. Calculate the second motion influence weight of each muscle node based on the influence of adjacent muscle nodes on each muscle node.

[0056] Besides being influenced by the bones to which they are attached, the smoothness of the muscle's mesh is also a crucial factor to consider, meaning the influence of adjacent muscle nodes must be taken into account. Adjacent muscle nodes are those connected to the current muscle node by an edge. In this case, the weight of the second motion influence depends only on the angle between the current muscle node and its surrounding adjacent muscle nodes, as shown in the following equation:

[0057] ;

[0058] in, It is the second motion influence weight of muscle node i on the influence of its neighboring muscle node j. It is a diagonal of the line connecting muscle node i and its adjacent muscle node j. It is the other diagonal of the line connecting muscle node i and its adjacent muscle node j, and m is the total number of adjacent muscle nodes.

[0059] Figure 4 This is a schematic diagram of the current muscle node v0 and its adjacent muscle nodes v1 to v5 provided in an embodiment of this application. The current muscle node v0 is affected by its adjacent muscle nodes v1 to v5. The diagonal of the line connecting the current muscle node v0 and its adjacent muscle node v1 is... and The diagonal of the line connecting the current muscle node v0 and its adjacent muscle node v2 is... and And so on, a total of 10 diagonals can be obtained: , , , , , , , , and After performing the cot operation on each of the 10 diagonals, the average is calculated to obtain the second motion influence weight of muscle node v0.

[0060] according to Figure 4 It can be seen that the diagonal angles are mostly between 0 and 90 degrees, while the cot function is a monotonically decreasing function between 0 and 180 degrees. If we assume that the five points v1 to v5 can enclose a relatively smooth surface, and v0 is the point to be adjusted, then when v0 is far from this smooth surface compared to when it is on the same surface, the diagonal angle value will be larger. In this case, the weight of the second motion influence will decrease, meaning the point that needs adjustment is deviating from the surface; conversely, if it is close to the same surface, the weight of the second motion influence will increase, meaning the original point needs less adjustment.

[0061] S140. Calculate the comprehensive weight of each muscle node based on the first motion influence weight and the second motion influence weight.

[0062] Optionally, the weights of the first and second movements can be weighted and summed to obtain the total weight for each muscle node, as shown in the following formula:

[0063] ;

[0064] in, It is the combined weight of the influence of the attached bone and the adjacent muscle node j on muscle node i. The weight of the first motion influence of muscle node i. This represents the second motion influence weight of muscle node i on its neighboring muscle node j. The weight for "weighted summation" in this step is 0.5, but it can be any other pre-calibrated value.

[0065] S150. Based on the comprehensive weight of the current muscle node and the positions of the current muscle node and its adjacent muscle nodes, calculate the Laplace coordinates of the current muscle node at the next moment.

[0066] See the following formula:

[0067] ;

[0068] in, It is the Laplace coordinate of muscle node i at the next time step. It is the overall weight of muscle node i. It is the set of adjacent muscle nodes of muscle node i. These are the absolute coordinates of muscle node i. These are the absolute coordinates of muscle node j.

[0069] Figure 5 This is a schematic diagram illustrating the changes in the position of muscle nodes provided in an embodiment of this application. Figure 5 The left side represents the muscle nodes v0~v7 at the current moment. Figure 5 The right side is a schematic diagram showing the change in the position of muscle node v0 to Δv0 at the next moment. Similarly, the angle at the current moment is... and The angle at the next moment is and .

[0070] Optionally, after calculating the Laplace coordinates of the current muscle node at the next time step, adjust the current muscle node to the Laplace coordinates and update the edges connecting the current muscle node to its adjacent muscle nodes. See [link to Laplace coordinates]. Figure 5 A schematic diagram showing the connection between muscle node v0 on the left and muscle node Δv0 on the right.

[0071] The Laplace coordinates (Δv0) of the current node are fixed. The Laplace coordinates of adjacent muscle nodes (e.g., muscle node v1) at the next time step are calculated, and their positions are adjusted and edges are updated. At the next time step, the adjacent nodes of muscle node v1 are Δv0, v2, v5, v6, and v7. Based on these adjacent nodes and the bone nodes to which muscle node v1 is attached, the comprehensive weight of muscle node v1 is calculated, and then the changed Laplace coordinates of muscle node v1 are calculated. This process is repeated until the Laplace coordinates of all muscle nodes are updated. It should be noted that for muscle nodes attached to bones, they need to move synchronously according to the absolute position coordinates of the bones; that is, muscle nodes attached to bones move as the position of the attached bones moves. For muscle nodes not attached to bones, the method provided in this embodiment is used to move them according to their Laplace coordinates.

[0072] This process continues until the Laplace coordinates of all muscle nodes are updated. To verify and improve the transformation results, the effectiveness of this method for deformation needs to be quantified.

[0073] The degree of Laplace coordinate change of all muscle nodes is characterized by the energy function E;

[0074] Current muscle node The changed position is defined as a point. Similarly, the position of the adjacent muscle point after the change is defined as So the difference in the Laplace coordinates after deformation It can be represented as:

[0075] ;

[0076] in, It is the overall weight of muscle node i. It is the set of adjacent muscle nodes of muscle node i.

[0077] Using a vertex energy function To describe the degree of change in the Laplace coordinates of the vertex before and after the deformation:

[0078] ;

[0079] Expanding the above equation:

[0080] ;

[0081] in, It is the absolute coordinate of muscle node i at the next moment. E represents the absolute coordinates of muscle node j at the next moment; n represents the total number of muscle nodes within the defined region. If the value of E is greater than the set threshold, it indicates that the defined region is not naturally deformed.

[0082] The set threshold can be pre-calibrated, and the set thresholds for different regions can be the same or different. If the value of E is greater than the set threshold, it indicates that there is unnatural deformation such as excessive stretching or twisting. See [link to relevant documentation]. Figure 6 It is necessary to reset the weights of the first and second motion influence weights and the "weighted sum" of the weights, and update the positions of each muscle node.

[0083] This embodiment applies to situations where the geometric model (mesh nodes) of muscles and bones changes accordingly when the human body's posture changes. See also Figure 7 When a person changes from a standing to a sitting posture, the hip muscles, such as the gluteus maximus, are among the parts of the body that undergo the most dramatic changes in shape: from relatively flat to severely compressed and spread out to the sides. The gluteus maximus (blue part) undergoes changes in position and shape under the combined influence of the skeleton and surrounding muscles. The position adjustment algorithm provided in this application, which combines the weights of the first and second motion effects, can accurately and efficiently complete the posture transformation of the gluteus maximus, improving the smoothness of the muscle nodes during the deformation process.

[0084] See Figure 8 The algorithm provided in this embodiment can better keep the point that needs to be smoothed (i.e. the current muscle node) and its adjacent muscle nodes on the same surface. For abnormal convex points on the muscle surface, after smoothing, they can be well integrated into the surrounding surface, thus reflecting the deformation characteristics of the muscle.

[0085] The embodiments of this application have the following technical effects:

[0086] 1. High fidelity: By constraining the Laplace coordinates and the shape of the muscle cross section, the generated muscle shape is natural and the local details are rich, effectively avoiding the distortion and volume loss problems of traditional skinning technology.

[0087] 2. High efficiency and automation: It enables automatic muscle generation from end to end, minimizing the workload of manual adjustments, greatly improving the development efficiency of digital human body models and shortening the development cycle.

[0088] 3. Anatomical rationality: Weight mapping and energy constraints based on anatomical knowledge make muscle deformation (such as stretching, contraction, and bulging) more consistent with real biomechanical behavior.

[0089] 4. Technological Integration and Innovation: By deeply integrating Laplace deformation from computer graphics with biomechanics and anatomy, a novel and reliable solution is provided for the specific problem of generating poses in digital human models.

[0090] This embodiment provides an electronic device, see [link / reference] Figure 9 It includes at least one processor 301 and a memory 302 communicatively connected to at least one of the processors 301;

[0091] The memory 302 stores instructions that can be executed by at least one of the processors 301, which enable at least one of the processors 301 to perform the above-described digital human muscle model pose generation method, thus having at least the same advantages as the above-described method.

[0092] Optionally, the electronic device also includes interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The components are interconnected using different buses and can be mounted on a common motherboard or otherwise installed as needed. The processor can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI (Graphical User Interface) on an external input / output device (such as a display device coupled to the interface). In other embodiments, multiple processors can be used with multiple memories, and / or multiple buses can be used with multiple memories, if desired. Similarly, multiple electronic devices (e.g., as a server array, a group of blade servers, or a multiprocessor system) can be connected, each providing some of the necessary operations.

[0093] The memory 302, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the digital human muscle model pose generation method in this embodiment. The processor 301 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 302, thereby realizing the aforementioned digital human muscle model pose generation method.

[0094] The memory 302 primarily includes a program storage area and a data storage area. The program storage area stores the operating system and at least one application program required for a given function; the data storage area stores data created based on terminal usage. Furthermore, the memory 302 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 302 may further include memory remotely configured relative to the processor, which can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0095] The electronic device may also include an input device 303 and an output device 304. The processor 301, memory 302, input device 303, and output device 304 may be connected via a bus or other means.

[0096] Input device 303 can receive input digital or character information, and output device 304 may include a display device, an auxiliary lighting device (e.g., an LED), and a haptic feedback device (e.g., a vibration motor). The display device may include, but is not limited to, a liquid crystal display (LCD), a light-emitting diode (LED) display, and a plasma display. In some embodiments, the display device may be a touchscreen.

[0097] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0098] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for generating poses of a digital human muscle model, characterized in that, include: High-precision CT scans are used to acquire tomographic images of the human body in its initial posture, and geometric models of the human skeleton and muscles are obtained by reverse modeling based on the tomographic images; the geometric model includes skeletal nodes and muscle nodes, as well as the connection relationships between the nodes; Based on each muscle node and the bone node to which it is attached, calculate the weight of the first motion influence of the attached bone on each muscle node. Calculate the second motion influence weight of each muscle node based on the influence of adjacent muscle nodes on each muscle node; Calculate the comprehensive weight of each muscle node based on the first motion influence weight and the second motion influence weight; Based on the overall weight of the current muscle node and the positions of the current muscle node and its adjacent muscle nodes, calculate the Laplace coordinates of the current muscle node at the next moment.

2. The method for generating poses of a digital human muscle model according to claim 1, characterized in that, In the geometric model, the skeleton is a rigid triangular mesh model, and the muscles are deformable triangular mesh models. After obtaining the geometric model of human bones and muscles through reverse modeling based on the tomographic image, the method further includes: generating a list of bone-muscle associations, which includes the bone node to which each muscle node is attached, and the absolute coordinates of each bone node.

3. The method for generating poses of a digital human muscle model according to claim 2, characterized in that, Based on each muscle node and the bone node it is attached to, calculate the weight of the first motion influence of the attached bone on each muscle node, including: For muscle nodes attached to skeletal nodes, the weight of the first motion influence is set to be greater than the set value; For muscle nodes not attached to skeletal nodes, the weight of the first motion influence is determined based on the weight distribution characteristics of the muscle spindle structure.

4. The method for generating poses of a digital human muscle model according to claim 3, characterized in that, The weight of the first motion influence is determined based on the weight distribution characteristics of the muscle spindle structure, including: ； ； ； in, The straight-line distance from the endpoint of the muscle-bone connection is The first movement at the muscle node affects the weight. for The radius of the approximately circular cross-section of the muscle. Let be the straight-line distance from the muscle node to the endpoint of the muscle-bone connection. Total muscle length This is a function related to the weight distribution of muscles; The coefficients are related to muscle type and current condition.

5. The method for generating poses of a digital human muscle model according to claim 1, characterized in that, Based on the influence of adjacent muscle nodes on each muscle node, the second motion influence weight of each muscle node is calculated, including: ； in, It is the second motion influence weight of muscle node i on the influence of its neighboring muscle node j. It is a diagonal of the line connecting muscle node i and its adjacent muscle node j. It is the other diagonal of the line connecting muscle node i and its adjacent muscle node j, and m is the total number of adjacent muscle nodes.

6. The method for generating poses of a digital human muscle model according to claim 1, characterized in that, Based on the first motion influence weight and the second motion influence weight, calculate the comprehensive weight for each muscle node, including: The weighted sum of the first and second motion influence weights is used to obtain the total weight of each muscle node.

7. The method for generating poses of a digital human muscle model according to claim 1, characterized in that, Based on the overall weight of the current muscle node, and the positions of the current muscle node and its adjacent muscle nodes, calculate the Laplace coordinates of the current muscle node at the next time step, including: ； in, It is the Laplace coordinate of muscle node i at the next time step. It is the overall weight of muscle node i. It is the set of adjacent muscle nodes of muscle node i. These are the absolute coordinates of muscle node i. These are the absolute coordinates of muscle node j.

8. The method for generating poses of a digital human muscle model according to claim 7, characterized in that, After calculating the Laplace coordinates of the current muscle node at the next time step, the following is also included: Adjust the current muscle node to the Laplace coordinates and update the edges connecting the current muscle node to its adjacent muscle nodes; Fix the Laplace coordinates of the current node, calculate the Laplace coordinates of the adjacent muscle nodes at the next time step and adjust their positions and update the edges, and so on, until the Laplace coordinates of all muscle nodes are updated.

9. The method for generating poses of a digital human muscle model according to claim 8, characterized in that, This process continues until the Laplace coordinates of all muscle nodes are updated, and then includes: The degree of Laplace coordinate change of all muscle nodes is characterized by the energy function E; ; in, It is the absolute coordinate of muscle node i at the next moment. is the absolute coordinate of muscle node j at the next moment; n is the total number of muscle nodes within the defined region; If the value of E is greater than the set threshold, it indicates that the set area is an unnatural deformation.

10. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to at least one of the processors; The memory stores instructions executable by at least one of the processors, which are executed by at least one of the processors to enable the at least one of the processors to perform the digital human muscle model pose generation method according to any one of claims 1-9.

Images