Virtual clothes simulation method, system, electronic device and storage medium

Abstract

The application provides a virtual clothes simulation method and system, an electronic device and a storage medium, and the method is as follows: physiological data of a user is collected; an interest area of a virtual clothes is screened out through the physiological data; different simulation precision to-be-simulated cloth vertices on the virtual clothes are screened out according to the interest area; and corresponding physical simulation strategies are respectively applied to the different simulation precision to-be-simulated cloth vertices. In the scheme, the interest area of the virtual clothes is screened out through the physiological data, then the different simulation precision to-be-simulated cloth vertices on the virtual clothes are screened out from the interest area, and finally the corresponding physical simulation strategies are respectively applied to the different simulation precision to-be-simulated cloth vertices, so that the unified high-precision physical solution does not need to be performed on all cloth vertices, thereby reducing the calculation cost and ensuring the real-time frame rate.

Description

Technical Field

[0001] This invention relates to the field of virtual reality technology, specifically to a virtual clothing simulation method, system, electronic device, and storage medium. Background Technology

[0002] In virtual reality (VR) and real-time computer graphics, creating realistic physical simulations of clothing on virtual characters is a computationally extremely expensive task.

[0003] Currently, the common approach is to perform a unified high-precision physics solution for all control vertices of the virtual clothing, and then simulate the cloth physics. However, this method has too high a computational cost and cannot guarantee a real-time frame rate in multi-person VR scenes. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a virtual clothing simulation method, system, electronic device, and storage medium to solve the problems of excessive computational overhead in existing simulation methods.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] The first aspect of this invention discloses a virtual clothing simulation method, the method comprising:

[0007] Collect users' physiological data;

[0008] The physiological data is used to filter out areas of interest for virtual clothing;

[0009] Based on the region of interest, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected;

[0010] Appropriate physical simulation strategies are applied to the vertices of the cloth to be simulated for different simulation accuracies.

[0011] Preferably, the interest areas of the virtual clothing are selected based on the physiological data, including:

[0012] The comprehensive interest value of each fabric vertex of the virtual garment is calculated using the physiological data.

[0013] The fabric vertices whose overall interest value is greater than the interest value threshold are selected to obtain the region of interest.

[0014] Preferably, calculating the comprehensive interest value of each fabric vertex of the virtual garment using the physiological data includes:

[0015] Using the physiological data, the rate of change of muscle stiffness at each fabric vertex of the virtual garment is calculated, as well as the global excitation weight of breathing on the region of interest is calculated.

[0016] A breadth-first search is performed on the fabric vertices where the rate of change of muscle stiffness is greater than a rate threshold to identify the force-generating muscle groups.

[0017] The comprehensive interest value of the fabric vertex is calculated by using the global excitation weight of the breathing on the region of interest and the Gaussian weight of the exerting muscle group on the fabric vertex.

[0018] Preferably, filtering out fabric vertices of different simulation precision on the virtual garment based on the region of interest includes:

[0019] Construct a basic bounding box for the region of interest;

[0020] The basic bounding box is expanded to obtain the final bounding box;

[0021] Based on the final bounding box, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected.

[0022] Preferably, the basic bounding box is expanded to obtain the final bounding box, including:

[0023] Calculate the expansion distance of the bounding box in each axis;

[0024] The basic bounding box is expanded using the expansion distance to obtain the final bounding box.

[0025] Preferably, the fabric vertices to be simulated with different simulation accuracies include: fabric vertices to be simulated with a first accuracy, a second accuracy, and a third accuracy; the first accuracy is higher than the second accuracy, and the second accuracy is higher than the third accuracy;

[0026] Appropriate physical simulation strategies are applied to the vertices of the cloth to be simulated at different simulation accuracies, including:

[0027] Perform a complete constraint projection on the vertices of the cloth to be simulated with the first precision.

[0028] Apply stretch constraints and body collision to the vertices of the fabric to be simulated for the second precision.

[0029] Skip the physical calculations for the third precision of the fabric vertices to be simulated.

[0030] A second aspect of this invention discloses a virtual clothing simulation system, the system comprising:

[0031] The data acquisition unit is used to collect the user's physiological data;

[0032] The first screening unit is used to filter out the areas of interest for the virtual clothing based on the physiological data.

[0033] The second filtering unit is used to filter out the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions based on the region of interest.

[0034] The simulation unit is used to apply corresponding physical simulation strategies to the vertices of the cloth to be simulated with different simulation accuracies.

[0035] Preferably, the first screening unit includes:

[0036] The calculation module is used to calculate the comprehensive interest value of each fabric vertex of the virtual garment using the physiological data;

[0037] The first filtering module is used to filter out the fabric vertices whose comprehensive interest value is greater than the interest value threshold, so as to obtain the region of interest.

[0038] A third aspect of the present invention discloses an electronic device, comprising: a processor and a memory, wherein the processor and the memory are connected via a bus; wherein the processor is used to call and execute a program stored in the memory; and the memory is used to store the program, the program being used to implement the virtual clothing simulation method disclosed in the first aspect of the present invention.

[0039] A fourth aspect of the present invention discloses a storage medium storing computer-executable instructions for executing the virtual clothing simulation method disclosed in the first aspect of the present invention.

[0040] Based on the above embodiments of the present invention, a virtual clothing simulation method, system, electronic device, and storage medium are provided. The method involves: collecting the user's physiological data; filtering out regions of interest (ROIs) for the virtual clothing using the physiological data; filtering out fabric vertices of different simulation accuracies on the virtual clothing based on the ROIs; and applying corresponding physical simulation strategies to the fabric vertices of different simulation accuracies. In this solution, the ROIs for the virtual clothing are filtered out using physiological data, and then fabric vertices of different simulation accuracies on the virtual clothing are filtered out from the ROIs. Finally, corresponding physical simulation strategies are applied to the fabric vertices of different simulation accuracies. This eliminates the need for a uniform high-precision physical solution for all fabric vertices, thereby reducing computational overhead and ensuring real-time frame rates. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0042] Figure 1 A flowchart illustrating a virtual clothing simulation method provided in an embodiment of the present invention;

[0043] Figure 2 A flowchart for filtering out regions of interest provided in an embodiment of the present invention;

[0044] Figure 3 This is a flowchart for filtering out the vertices of the cloth to be simulated, provided in an embodiment of the present invention.

[0045] Figure 4 A structural block diagram of a virtual clothing simulation system provided in an embodiment of the present invention; Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0048] In virtual reality (VR) and real-time computer graphics, realistically simulating the physics of clothing worn by virtual characters has always been a computationally extremely demanding task. This is especially true in large-scale, multi-person interactive VR scenarios (such as virtual gyms), where each frame is highly sensitive to latency and computing resources are severely limited.

[0049] Currently, the common approach is to perform a unified high-precision physics solution for all control vertices of the virtual clothing, and then simulate the cloth physics. However, this method has too high a computational cost and cannot guarantee a real-time frame rate in multi-person VR scenes.

[0050] To address this, this invention proposes a virtual clothing simulation method, system, electronic device, and storage medium. It uses physiological data to filter out regions of interest in the virtual clothing, then uses these regions to filter out fabric vertices of different simulation accuracies on the virtual clothing. Finally, it applies corresponding physical simulation strategies to the fabric vertices of different simulation accuracies, eliminating the need for uniform high-precision physical solutions for all fabric vertices, thereby reducing computational overhead and ensuring real-time frame rates.

[0051] See Figure 1 The flowchart illustrates a virtual clothing simulation method provided by an embodiment of the present invention, the method comprising:

[0052] Step S101: Collect the user's physiological data.

[0053] In the specific implementation step S101, various sensors and devices are used to collect the user's physiological data, which includes at least: electromyography amplitude (raw value denoted as...). Heart rate (referred to as HR) raw (t)), chest expansion (denoted as E) raw (t)), rotational quadruples of 24 joints and world coordinates of the root skeleton.

[0054] Where i and j represent rows and columns, raw means raw, and t means timestamp.

[0055] Specifically, sensors are placed in the main muscle groups of the inner layer of the smart clothing, and these sensors return 16 A 16-dimensional pressure array (i.e., electromyographic amplitude), the raw values ​​returned by the sensor are denoted as... .

[0056] The chest strap integrates an ECG or bioelectrical impedance sensor to output heart rate and chest expansion.

[0057] The rotation quadruples of 24 joints and the world coordinates of the root skeleton are calculated using VR headset, controllers and limb IMUs to drive the virtual avatar skeleton.

[0058] It should be noted that in subsequent applications, the electromyographic amplitude... Transformed into muscle stiffness value H(v,t) through skin mapping; Heart rate HR raw (t) is used to adjust the maximum respiratory rate (denoted as RR) during strenuous exercise. max Thoracic expansion E raw (t) is used to calculate the instantaneous respiratory rate RR(t) and respiratory phase. (t); The joint rotation quadruple is used to calculate the maximum value v of the instantaneous velocity scalar at the vertex. maxThe quaternion of root skeleton world coordinates and joint rotation is used to determine the absolute position of the virtual avatar in the VR space.

[0059] Preferably, in practical applications, after collecting the user's physiological data, the collected physiological data needs to be filtered and time-aligned, specifically involving processing such as de-jittering, noise reduction, time alignment, and missing value imputation.

[0060] Shake reduction and noise reduction: Electromyographic amplitude Perform median filtering with a window size of 3 (example only); for heart rate (HR)... raw (t) is filtered using a moving average to evaluate the thoracic expansion E. raw (t) is low-pass filtered to eliminate high-frequency noise caused by body movement, resulting in clean heart rate HR(t) and chest expansion E(t).

[0061] Time alignment: All sensor data is remapped to a uniform 120Hz logical frame through linear interpolation to eliminate sampling rate differences.

[0062] Missing value filling: If a sensor fails in a certain frame, the missing value is filled by interpolation between the previous valid frame and the default reference value, and the confidence level is marked.

[0063] Step S102: Select the areas of interest for the virtual clothing using physiological data.

[0064] In the specific implementation step S102, the regions of interest (ROI) of the virtual clothing are selected based on the collected physiological data.

[0065] Step S103: Select the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions based on the region of interest.

[0066] In the specific implementation step S103, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected according to the region of interest. The fabric vertices to be simulated with different simulation precisions include: fabric vertices with first precision, second precision and third precision; the first precision is higher than the second precision, and the second precision is higher than the third precision.

[0067] It should be noted that the first precision is high precision (HIGH), the second precision is medium precision (MEDIUM), and the third precision is low precision (LOW).

[0068] Step S104: Apply corresponding physical simulation strategies to the vertices of the cloth to be simulated with different simulation accuracies.

[0069] In the specific implementation of step S104, corresponding physical simulation strategies are applied to the cloth vertices to be simulated for different simulation accuracies.

[0070] In this embodiment of the invention, the region of interest of the virtual clothing is selected by physiological data, and then the vertices of the fabric to be simulated with different simulation precisions on the virtual clothing are selected by the region of interest. Finally, the corresponding physical simulation strategies are applied to the fabric vertices to be simulated with different simulation precisions. It is not necessary to perform a uniform high-precision physical solution for all fabric vertices, thereby reducing the computational overhead and ensuring the real-time frame rate.

[0071] Regarding the above embodiments of the present invention Figure 1 The region of interest involved in step S102, see [link / reference]. Figure 2 This illustrates a flowchart for filtering regions of interest according to an embodiment of the present invention. Figure 2 Includes the following steps:

[0072] Step S201: Calculate the comprehensive interest value of each fabric vertex of the virtual garment using physiological data.

[0073] In the specific implementation step S201, physiological data is used to calculate the rate of change of muscle stiffness at each fabric vertex of the virtual garment (denoted as ΔH(v,t)), and to calculate the global excitation weight of breathing on the region of interest (denoted as w). breath (t)).

[0074] Specifically, the filtered After flattening and normalizing, a one-dimensional normalized pressure vector is obtained. Then, by using linear blending skinning to map each cloth vertex v of the virtual avatar body mesh, the muscle hardness value of the cloth vertex (denoted as H(v,t)) is obtained, which is specifically calculated by formula (1).

[0075] (1);

[0076] Wherein, H(v,t) is the muscle stiffness value of the cloth vertex v at time t, which is dimensionless and ranges from 0 to 1; Let V be the skin weight of the fabric vertex v affected by the i-th sensor, and let its sum be 1; Let be the normalized reading of the i-th sensor at time t, ranging from 0 to 1.

[0077] In addition, peak detection and zero-crossing analysis were performed on the thoracic expansion E(t) to output the instantaneous respiratory rate RR(t) and respiratory phase. (t) (the peak of inhalation is 0, and the peak of exhalation is π).

[0078] It should be noted that only muscles that are "hardening" can exert a significant dynamic push on clothing. Therefore, the time gradient of muscle hardness value is calculated by formula (2) to obtain the rate of change of muscle hardness at the apex of the fabric, ΔH(v,t).

[0079] (2);

[0080] In formula (2), ΔH(v,t) is the rate of change of muscle stiffness at the fabric vertex v at time t, in units of stiffness / second; Δt is the time interval between two logical frames (i.e., 1 / 120 seconds).

[0081] It should be noted that the thoracic region should be primarily activated by respiratory rate, and heart rate can also help expand the ROI of the chest. The global activation weight w of breathing on the region of interest can be calculated using formula (3). breath (t).

[0082] (3);

[0083] In formula (3), w breath (t) represents the global excitation weight of respiration on the region of interest, ranging from 0 to 1; RR(t) represents the instantaneous respiratory rate (breaths per minute); RR rest This refers to the resting respiratory rate; RR max This is the maximum respiratory rate during strenuous exercise.

[0084] For the rate of change of muscle stiffness ΔH(v,t) greater than the rate threshold (denoted as τ) grad A breadth-first search is performed on the fabric vertices to identify the force-generating muscle group G. k .

[0085] It should be noted that isolated activated vertices are meaningless; only spatially connected vertex groups exceeding a threshold should be identified as the entire muscle group exerting force. The velocity threshold τ grad It is the discrimination threshold applied to ΔH(v,t).

[0086] Therefore, for ΔH(v,t)>τ grad Perform a breadth-first search (BFS) on the cloth vertices to find all connected components, discarding those with fewer than N vertices. min (For example, 50) connected components, and each remaining connected component is denoted as a muscle group G. k .

[0087] It should be noted that k is an integer index starting from 1, used to distinguish the detected muscle groups involved in the exertion of force, G. k Essentially, it's a subset of cloth vertices: it contains a group of interconnected cloth vertices in three-dimensional space, where the rate of change of muscle stiffness at each vertex exceeds the rate threshold τ. grad .

[0088] The global incentive weight w of the region of interest is determined by breathing. breath (t) and the force-generating muscle group G kThe Gaussian weights of the cloth vertices are used to calculate the overall interest value (denoted as I(v)) of the cloth vertices.

[0089] Specifically, the muscle force group, cardiopulmonary drive and optional gaze point are fused into a 0-1 continuous scalar field I(v) to avoid rigid binary boundaries, that is, the comprehensive interest value I(v) of the cloth vertex is calculated by formula (4).

[0090] (4);

[0091] In formula (4), I(v) is the comprehensive interest value of the fabric vertex v, ranging from 0 to 1; For the force-generating muscle group G k The 3D Gaussian weights for the fabric vertices are set with a weight of 1 at the center of the muscle group, decreasing outwards; α, β, and γ are weight coefficients, representing the contributions of the muscle, breathing, and fixation terms, respectively, which can be set manually or obtained through machine learning. For the indicator function, when v belongs to the chest vertex set V chest The value is 1 if it is true, and 0 otherwise. This is the interest value assigned by the user's gaze direction (optional). If gaze-driven is enabled, it is the decay weight of the vertex near the gaze point; otherwise, it is 0.

[0092] Step S202: Filter out the fabric vertices whose overall interest value is greater than the interest value threshold to obtain the region of interest.

[0093] In the specific implementation step S202, samples with a comprehensive interest value I(v) greater than the interest value threshold (denoted as τ) are selected. act The cloth vertices are used to obtain the region of interest, which is a set of vertices (denoted as V). roi In other words, the region of interest V roi It includes "I(v)>τ" act The apex of the fabric.

[0094] The above embodiments of the present invention Figure 2 This is an explanation of how to obtain areas of interest.

[0095] Regarding the above embodiments of the present invention Figure 1 For the cloth vertices to be simulated involved in step S103, see [link / reference]. Figure 3 This illustrates a flowchart of filtering out the vertices of the cloth to be simulated, provided by an embodiment of the present invention. Figure 3 Includes the following steps:

[0096] Step S301: Construct a basic bounding box for the region of interest.

[0097] In the specific implementation step S301, the region of interest V is traversed. roiGiven the world coordinates (x, y, z) of all cloth vertices, find the minimum point P. min and the maximum point P max .

[0098] P min =(min x i ,min y i ,min z i ), P max =(max x i ,max y i ,max z i The basic bounding box is [P]. min , P max ].

[0099] Step S302: Expand the basic bounding box to obtain the final bounding box.

[0100] It should be noted that the basic bounding box cannot adapt to the movement lag of the fabric apex during muscle expansion and high-speed arm swings, so dynamic margins based on speed and physiological state must be introduced.

[0101] In the specific implementation of step S302, the expansion distance δ of the bounding box in each axis is calculated by formula (5) (in meters or centimeters).

[0102] (5);

[0103] In formula (5), The base extension, taking into account fabric thickness and vertex density, is typically taken as 0.02m; This is the velocity expansion factor, typically taken as 0.03s to 0.05s; For region of interest V roi The maximum instantaneous velocity scalar value of all cloth vertices; The logical frame time interval (1 / 120 seconds); The extra expansion coefficient for respiration, per unit length (e.g., 0.05m). The global incentive weight for breathing in the region of interest.

[0104] Using the extended distance δ to the basic bounding box [P min , P max Expanding the bounding box, we get the final bounding box as [P]. min -δ,P max +δ].

[0105] Step S303: Based on the final bounding box, filter out the vertices of the fabric to be simulated on the virtual clothing with different simulation precision.

[0106] It should be noted that the cloth vertices that need to be simulated with high precision are selected based on the final bounding box, and a transition zone is formed by neighborhood expansion to ensure that the subsequent physics simulation will not be torn due to abrupt changes in precision.

[0107] First, the high-precision vertex set V high V is used to store the vertices of the cloth to be simulated with first-precision precision, and the vertex set of the boundary transition band. medium Used to store the cloth vertices to be simulated with second precision.

[0108] In the specific implementation step S303, each fabric vertex P of the virtual clothing fabric mesh is processed. clouth Determine the vertex P of the cloth. clouth Do the world coordinates satisfy "P"? min -δ≦P clouth ≦P max +δ” (components satisfy simultaneously), that is, to determine the cloth vertex P clouth Are the world coordinates within the final bounding box? [P] min -δ, P max +δ] within. If the aforementioned conditions are met, then the cloth vertex P is... clouth Put V in medium middle.

[0109] It should be noted that in physical simulations, if the high-precision region and the low-precision region are directly adjacent, the connecting spring will experience a huge energy difference due to the different solution accuracies at both ends, resulting in severe vibration. Therefore, it is necessary to extend it outward by at least one triangular face.

[0110] Traverse all the triangular edges of the virtual clothing fabric mesh. If one end of an edge has a fabric vertex belonging to V... medium The vertex of the fabric at the other end does not belong to V. medium Then "not belonging to V" medium "Cloth vertex" is forcibly added to V high The aforementioned operation is performed 1 to 2 times to generate the final V. high and V medium .

[0111] V high The middle part contains the first-precision cloth vertex to be simulated, V medium The middle part contains the cloth vertices to be simulated with the second precision, and the remaining cloth vertices contain the cloth vertices to be simulated with the third precision.

[0112] Using the above method, the fabric vertices can be divided into the following three regions:

[0113] High-precision region, i.e., the first precision (HIGH) of the cloth vertex to be simulated: perform complete high-precision solution;

[0114] Lightweight solution is performed on the medium precision region, i.e., the second precision (MEDIUM) of the cloth vertex to be simulated: the boundary extension zone.

[0115] Low-precision regions, i.e., third-precision (LOW) fabric vertices to be simulated: only skinning is performed, without physical solution.

[0116] The above embodiments of the present invention Figure 3 This is an explanation of how to select the vertices of the cloth to be simulated.

[0117] In some embodiments, corresponding physical simulation strategies are applied to the vertices of the cloth to be simulated with different simulation accuracies. Here, stiffness mapping and the construction of the breathing external force field are explained first.

[0118] Stiffness mapping: The clothing on the surface of the area where the muscle hardens due to exertion should appear tighter, that is, the local fabric stiffness increases. Specifically, stiffness mapping is performed using formula (6).

[0119] (6);

[0120] In formula (6), The elastic coefficient of the fabric spring near the apex v of the fabric attached to the body; It provides basic stiffness to ensure the fabric maintains its normal shape; H(v,t) is the muscle stiffness influence index, a positive number that controls the contribution of stiffness to stiffness; H(v,t) is the muscle stiffness value of cloth vertex v in the current frame; I(v) is the overall interest value of cloth vertex v, used to ensure that stiffness changes are significant only within the ROI, and naturally decay in the transition zone.

[0121] Construction of the external force field for breathing: The external force field for breathing is constructed by simulating the rhythmic lifting and retraction of clothing by the chest cavity, specifically through formula (7).

[0122] (7);

[0123] In formula (7), The simulated external force for breathing is applied to the apex v of the chest fabric; This refers to the amplitude of the inhalation force, which controls the magnitude of the external thrust during inhalation. This is the baseline value for expiratory force, which can be negative (inward pull) or a small positive value; (t) represents the respiratory phase; Let v be the direction of the normal to the body surface at the vertex of the fabric, ensuring that the direction of the force is either pushing outward or pulling inward along the surface normal.

[0124] Based on the above stiffness mapping and breathing external force field construction, the specific implementation of applying corresponding physical simulation strategies to the cloth vertices to be simulated with different simulation accuracies is as follows: perform complete constraint projection on the cloth vertices to be simulated with the first accuracy; perform stretch constraints and body collision on the cloth vertices to be simulated with the second accuracy; skip the physical calculation of the cloth vertices to be simulated with the third accuracy.

[0125] Specifically, a complete constraint projection is performed on the first-precision simulated cloth vertices: a position-based dynamics (PBD) or "mass-spring model" is used to perform a complete constraint projection on the first-precision simulated cloth vertices, including tension constraints, bending constraints, self-collision, and body collision, iterating 5 to 8 times. The spring stiffness is determined using formula (6). External forces include gravity and the force in formula (8). .

[0126] It should be noted that constrained projection refers to a position correction mechanism in the fabric physics simulation process. Traditional fabric simulation usually solves the motion of particles based on force analysis, while the Position-Based Dynamics (PBD) method takes a different approach: first, it predicts the initial position of the fabric particles through numerical integration, and then iteratively corrects the predicted position using various constraint equations to make the particle position satisfy the physical properties of the fabric (such as instretchability, bending resistance, etc.). This correction process is called "constrained projection".

[0127] The core advantage of constrained projection lies in its ability to directly control particle displacement, avoiding numerical instability caused by ill-conditioned stiffness matrices in force-based methods, while maintaining the conservation of momentum and angular momentum of the system. In this invention, a complete constrained projection is performed on the high-precision region, including iterative solutions for various constraint types such as tension constraints, bending constraints, self-collision, and body collision, to ensure the physical realism and stability of the cloth simulation.

[0128] Apply stretch constraints and body collision to the vertices of the cloth to be simulated at the second precision level: Only apply stretch constraints and body collision to the vertices of the cloth to be simulated at the second precision level, iterating 1-2 times. Anchor the edge vertices to the output positions of the high-precision region to prevent them from detaching.

[0129] It should be noted that tension constraints are one of the most basic constraint types in cloth physics simulations, used to limit the distance variation between adjacent particles in the cloth mesh. In a position-based dynamics framework, tension constraints limit the distance between adjacent particles to the allowable range of the initial rest length through constraint equations, thereby simulating the physical property of cloth being "bendable and tensile" (i.e., cloth can bend freely to produce wrinkles, but has high stiffness in the tensile direction).

[0130] Body collision refers to the collision detection and response handling between cloth and the surface of a virtual character's body. In real-time cloth simulation, if the penetration between clothing and the body is not effectively handled, the vertices of the clothing may sink into the body, resulting in severe clipping and compromising visual realism. A common approach is to abstract the human body surface into simplified collision objects such as capsules, spheres, or mesh-enclosed bodies. Cloth particles are then subjected to collision detection with these collision objects, and when penetration is detected, position correction is applied, pushing the particles back to the outside of the body surface.

[0131] Third-precision cloth vertices to be simulated: Skip the physical calculations for third-precision cloth vertices to be simulated, and directly use dual quaternion skinning (DQS) driven by skeletal motion with zero additional CPU overhead.

[0132] Physics-based computation (or physics simulation solution) refers to the process of numerically solving the equations of motion of cloth particles to update the position and velocity of each particle in each frame. In traditional cloth simulation methods, physics-based computation typically involves constructing a particle-spring model or a position-based dynamic model, and simulating the deformation behavior of the cloth by iteratively solving various constraints such as tension constraints, bending constraints, and collision constraints.

[0133] It should be noted that, taking the mass-spring model as an example, the motion equation of the mass in the high-precision region is shown in formula (8).

[0134] (8);

[0135] In formula (8), m is the mass of the particle; a is the acceleration of the particle; The resultant force of the spring is the spring stiffness inside, which is dynamically determined by the stiffness mapping formula given by formula (6); It is a damping force, directly proportional to the particle velocity, and in the opposite direction; The resultant force is the external force, including gravity and the force in formula (8). Wind power, etc.

[0136] After performing a complete constraint projection on the vertices of the cloth to be simulated with the first precision, and Substitute into formula (8) to perform numerical solution, update the position of the first precision cloth vertex to be simulated, and submit the updated first precision cloth vertex to the GPU for rendering in the VR headset.

[0137] The above is a detailed description of the present invention. One application scenario after applying this invention is as follows: In a large-scale virtual reality gym, a user enters the scene wearing a VR headset and a smart bodysuit integrating electromyography, pressure, heart rate, and respiratory sensors. Their virtual avatar is wearing workout clothes. When the user performs strength training, their muscles become engorged and hardened, and their heart rate and breathing increase. The virtual clothing displays a tight, supportive effect in real time at the muscle groups being used, and rises and falls in rhythm with breathing in the chest and abdomen, while maintaining low-overhead simulation in other areas. Through this invention, computing power is precisely allocated to physiologically active areas, achieving a balance between realism and performance in multi-person VR environments.

[0138] Corresponding to the virtual clothing simulation method provided in the above embodiments of the present invention, the present invention also provides a structural block diagram of a virtual clothing simulation system, which includes: a data acquisition unit 401, a first screening unit 402, a second screening unit 403, and a simulation unit 404;

[0139] The acquisition unit 401 is used to collect the user's physiological data.

[0140] The first screening unit 402 is used to filter out the areas of interest of the virtual clothing based on physiological data.

[0141] The second filtering unit 403 is used to filter out the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions according to the region of interest.

[0142] The simulation unit 404 is used to apply corresponding physical simulation strategies to the vertices of the cloth to be simulated with different simulation accuracies.

[0143] In the specific implementation, the cloth vertices to be simulated with different simulation accuracies include: cloth vertices to be simulated with first accuracy, second accuracy, and third accuracy; the first accuracy is higher than the second accuracy, and the second accuracy is higher than the third accuracy; the simulation unit 404 is specifically used to: perform a complete constraint projection on the cloth vertices to be simulated with first accuracy; perform stretch constraints and body collision on the cloth vertices to be simulated with second accuracy; and skip the physical calculations on the cloth vertices to be simulated with third accuracy.

[0144] Preferred, combined Figure 4 The content shown indicates that the first filtering unit 402 includes a calculation module and a first filtering module. The execution principle of each module is as follows:

[0145] The calculation module is used to calculate the comprehensive interest value of each fabric vertex of the virtual garment using physiological data.

[0146] In specific implementation, the calculation module is used to: calculate the rate of change of muscle stiffness at each fabric vertex of the virtual clothing using physiological data, and calculate the global excitation weight of breathing on the region of interest; perform a breadth-first search on fabric vertices with a rate of change of muscle stiffness greater than the rate threshold to find the muscle groups exerting force; and calculate the comprehensive interest value of the fabric vertices using the global excitation weight of breathing on the region of interest and the Gaussian weight of the muscle groups exerting force on the fabric vertices.

[0147] The first filtering module is used to filter out the fabric vertices whose comprehensive interest value is greater than the interest value threshold, so as to obtain the region of interest.

[0148] Preferred, combined Figure 4 As shown, the second filtering unit 403 includes a construction module, an extension module, and a second filtering module. The execution principle of each module is as follows:

[0149] The construction module is used to construct the basic bounding box for the region of interest.

[0150] Extension modules are used to extend the base bounding box to obtain the final bounding box.

[0151] Specifically, the extension module is used to: calculate the extension distance of the bounding box in each axis; and extend the basic bounding box using the extension distance to obtain the final bounding box.

[0152] The second filtering module is used to filter out the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions based on the final bounding box.

[0153] Preferably, the present invention also provides an electronic device, including: a processor and a memory, the processor and the memory being connected via a bus; wherein, the processor is used to call and execute a program stored in the memory; the memory is used to store the program, the program being used to implement the virtual clothing simulation method provided in the above method embodiments.

[0154] Preferably, the present invention also provides a storage medium storing computer-executable instructions for executing the virtual clothing simulation method provided in the above-described method embodiments.

[0155] In summary, the embodiments of the present invention provide a virtual clothing simulation method, system, electronic device, and storage medium. By filtering out the regions of interest of the virtual clothing through physiological data, and then filtering out the vertices of the fabric to be simulated with different simulation precisions from the regions of interest, the present invention applies corresponding physical simulation strategies to the vertices of the fabric to be simulated with different simulation precisions. This eliminates the need to perform a uniform high-precision physical solution for all fabric vertices, thereby reducing computational overhead and ensuring real-time frame rate.

[0156] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0157] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0158] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A virtual clothing simulation method, characterized in that, The method includes: Collect users' physiological data; The physiological data is used to filter out areas of interest for virtual clothing; Based on the region of interest, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected; Appropriate physical simulation strategies are applied to the vertices of the cloth to be simulated for different simulation accuracies.

2. The method according to claim 1, characterized in that, The physiological data is used to filter out areas of interest for virtual clothing, including: The comprehensive interest value of each fabric vertex of the virtual garment is calculated using the physiological data. The fabric vertices whose overall interest value is greater than the interest value threshold are selected to obtain the region of interest.

3. The method according to claim 2, characterized in that, The comprehensive interest value of each fabric vertex of the virtual garment is calculated using the physiological data, including: Using the physiological data, the rate of change of muscle stiffness at each fabric vertex of the virtual garment is calculated, as well as the global excitation weight of breathing on the region of interest is calculated. A breadth-first search is performed on the fabric vertices where the rate of change of muscle stiffness is greater than a rate threshold to identify the force-generating muscle groups. The comprehensive interest value of the fabric vertex is calculated by using the global excitation weight of the breathing on the region of interest and the Gaussian weight of the exerting muscle group on the fabric vertex.

4. The method according to claim 1, characterized in that, Based on the region of interest, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected, including: Construct a basic bounding box for the region of interest; The basic bounding box is expanded to obtain the final bounding box; Based on the final bounding box, the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions are selected.

5. The method according to claim 4, characterized in that, The basic bounding box is expanded to obtain the final bounding box, including: Calculate the expansion distance of the bounding box in each axis; The basic bounding box is expanded using the expansion distance to obtain the final bounding box.

6. The method according to claim 1, characterized in that, The fabric vertices to be simulated at different simulation accuracies include: first accuracy, second accuracy, and third accuracy; the first accuracy is higher than the second accuracy, and the second accuracy is higher than the third accuracy. Appropriate physical simulation strategies are applied to the vertices of the cloth to be simulated at different simulation accuracies, including: Perform a complete constraint projection on the vertices of the cloth to be simulated with the first precision. Apply stretch constraints and body collision to the vertices of the fabric to be simulated for the second precision. Skip the physical calculations for the third precision of the fabric vertices to be simulated.

7. A virtual clothing simulation system, characterized in that, The system includes: The data acquisition unit is used to collect the user's physiological data; The first screening unit is used to filter out the areas of interest for the virtual clothing based on the physiological data. The second filtering unit is used to filter out the vertices of the fabric to be simulated on the virtual clothing with different simulation precisions based on the region of interest. The simulation unit is used to apply corresponding physical simulation strategies to the vertices of the cloth to be simulated with different simulation accuracies.

8. The system according to claim 7, characterized in that, The first filtering unit includes: The calculation module is used to calculate the comprehensive interest value of each fabric vertex of the virtual garment using the physiological data; The first filtering module is used to filter out the fabric vertices whose comprehensive interest value is greater than the interest value threshold, so as to obtain the region of interest.

9. An electronic device, characterized in that, include: A processor and a memory are connected via a bus; wherein the processor is used to call and execute a program stored in the memory; The memory is used to store a program for implementing the virtual clothing simulation method as described in any one of claims 1-6.

10. A storage medium, characterized in that, The storage medium stores computer-executable instructions for performing the virtual clothing simulation method as described in any one of claims 1-6.

Images