Motion capture and evaluation method, device, equipment and medium

Abstract

The invention relates to a motion capture and evaluation method and device, equipment and a medium. The method comprises the following steps: acquiring motion capture data; repairing the motion capture data to obtain standardized motion data; obtaining a virtual character animation file based on the standardized action data and a virtual character skeleton model; performing multi-dimensional evaluation on the virtual character animation file to generate a comprehensive evaluation report; and adjusting the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file. By adopting the method, the motion capture animation can be automatically evaluated and corrected, and the effect of improving the working efficiency is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of animation production, and in particular relates to a motion capture and evaluation method, device, equipment and medium. Background Art

[0002] With the development of 3D animation production technology, motion capture 3D modeling has emerged. However, the current mainstream motion capture-to-animation pipeline still relies on a traditional framework: raw motion data is collected through optical markers or inertial sensors. After Butterworth low-pass filtering and linear interpolation to correct occlusions, joint rotation data is directly mapped to a standard skeleton. Inverse kinematics is used to adapt the virtual character's proportions, and the final output is an animation file that can be driven by the engine. Due to the lack of automation in the evaluation and correction stages, this mainstream motion capture-to-animation pipeline relies heavily on manual experience for evaluation and correction. Summary of the Invention

[0003] Based on this, it is necessary to provide a motion capture and evaluation method, device, equipment and medium that can evaluate animation quality and improve work efficiency in response to the above technical problems.

[0004] In a first aspect, the present application provides a motion capture and evaluation method, comprising:

[0005] Get motion capture data;

[0006] Repair the motion capture data to obtain standardized motion data;

[0007] Based on the standardized motion data and the virtual character skeleton model, a virtual character animation file is obtained;

[0008] Evaluate virtual character animation files from multiple dimensions and generate a comprehensive evaluation report;

[0009] Adjust the virtual character animation file based on the comprehensive evaluation report to obtain the final animation file.

[0010] Furthermore, the virtual character animation files are evaluated in multiple dimensions to generate a comprehensive evaluation report, including:

[0011] Based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal coordination analysis. The coordination evaluation report is used to characterize the temporal synchronization and spatial coordination efficiency of different body parts of the virtual character animation file.

[0012] Based on the motion trajectory in the virtual character animation file, a fluency evaluation report is obtained; the fluency evaluation report is used to characterize the action continuity of the motion trajectory;

[0013] Based on the joint angles of the virtual character animation file, the over-limit joint frames are extracted and a rationality assessment report is generated; the rationality assessment is used to characterize the biomechanical rationality of the action;

[0014] Based on the virtual character animation file and the virtual character's physical parameters, the center of mass projection is verified and a physical inertia evaluation report is generated; the physical inertia evaluation report is used to characterize the physical credibility of the action;

[0015] Integrate the coordination assessment report, fluency assessment report, rationality assessment report and physical inertia assessment report to generate a comprehensive assessment report.

[0016] Furthermore, based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal synergy analysis, including:

[0017] Based on the upper and lower limb movements of the avatar animation file, dynamic time warping is used to calculate the minimum path cumulative cost using the following formula to obtain the optimal alignment path:

[0018]

[0019] Among them, C(i,j) is the minimum path cumulative cost from the starting point to the current point, a i is the data of the i-th frame of the reference sequence, b j is the data of the jth frame of the target sequence, d(a i ,b j ) is the local distance of the quantity sequence in frame (i, j);

[0020] Based on the optimal alignment path, the average action delay is obtained by the following formula:

[0021]

[0022] Where a is the average action delay, N is the total length of the optimal alignment path, i k ,j k is the frame number of the reference sequence and target sequence corresponding to the kth step of the path;

[0023] Based on the average action delay, the timing deviation exceeding the limit frame is obtained; the timing deviation exceeding the limit frame indicates that the upper and lower limb movements are out of sync;

[0024] The dominant motion in the virtual character animation file is extracted through principal component analysis to obtain the motion contribution;

[0025] The abnormal dominant kinematic chain is obtained by comparing the movement contribution with the biomechanical standard, which is extracted from the biomechanical reference standard library.

[0026] Based on the timing deviation exceeding the limit frame and the abnormal dominant motion chain, the coordination assessment report is determined.

[0027] Furthermore, based on the motion trajectory in the virtual character animation file, a fluency evaluation report is obtained, including:

[0028] Based on the virtual character animation file, the joint angular acceleration is obtained by the second-order difference method;

[0029] Compare the joint angular acceleration with the threshold to obtain the acceleration exceeding limit frame;

[0030] Based on the Bezier curve fitting of the hand trajectory in the virtual character animation file, the curvature standard deviation is obtained using the following formula:

[0031] B(t)=(x(t),y(t))

[0032]

[0033] Among them, σ κ is the standard deviation of curvature, B(T) is the Bezier curve parameter equation of the hand trajectory, x(t), y(t) are the horizontal and vertical coordinates of the hand at time t, κ(t) is the curvature of the hand trajectory at time t, x′(t), y′(t) are the first-order derivatives of the hand trajectory at time t, x”(t), y”(t) are the second-order derivatives of the hand trajectory at time t, N is the total number of frames of the trajectory, κ i is the curvature value of the i-th frame, μ κ is the average value of curvature;

[0034] Compare the curvature standard deviation to the threshold to obtain the unnatural linear motion frame;

[0035] A smoothness evaluation report is obtained based on acceleration exceeding limit frames and unnatural linear motion frames.

[0036] Furthermore, the motion capture data is repaired to obtain standardized motion data, including:

[0037] De-noise the motion capture data to obtain the denoised motion data file;

[0038] Perform singular value decomposition on the denoised motion data file to obtain the rotation matrix and translation vector of the denoised motion data file to the virtual world coordinate system;

[0039] The joint data in the denoised motion data file is converted to a unified virtual world coordinate system based on the rotation matrix and translation vector to obtain a coordinate-aligned motion data file;

[0040] PCA reconstruction of the motion data file based on coordinate system alignment repairs missing data to obtain a complete motion data file;

[0041] The complete motion data file is band-pass filtered to obtain the respiratory layer signal; the respiratory layer signal is an independent chest cavity rise and fall signal;

[0042] Remove the respiratory layer signal from the complete motion data file to obtain the main motion layer signal; the main motion layer signal is pure limb movement data;

[0043] The respiratory layer signal and the main motion layer signal are resampled and aligned to obtain standardized motion data.

[0044] Furthermore, based on the standardized motion data and the avatar skeleton model, a avatar animation file is obtained, including:

[0045] Align the standard human skeleton structure with the character mesh model to obtain a skeleton-bound character model;

[0046] The vertex weights of the character model with bound bones are calculated using a heat diffusion algorithm to obtain the skeleton model of the virtual character; the vertex weights affect the degree of model deformation when the joints are bent;

[0047] Adjust the standardized motion data movement amplitude according to the virtual character skeleton model to obtain the initial calibration animation file;

[0048] Based on the initial calibration animation file and the virtual character skeleton model, the end effector position is solved by inverse kinematics to obtain the preliminary redirection animation file;

[0049] Based on the physical parameters of the virtual character, physical constraints are applied to the preliminary redirected animation file to obtain the constrained animation file;

[0050] Based on the virtual character's physical parameters, the center of mass of the constrained animation file is balanced to obtain a physically corrected animation file;

[0051] The key frame curve of the physically corrected animation file is smoothed to obtain the virtual character animation file.

[0052] Furthermore, after adjusting the virtual character animation file based on the comprehensive evaluation report to obtain the final animation file, the following steps are also included:

[0053] receiving preset art enhancement instructions; the art enhancement instructions are used to instruct stylization of the final animation file;

[0054] Based on the art enhancement instructions, adjust the key frames of the final animation file to obtain a preliminary stylized animation file;

[0055] Adjust the initial stylized animation file in and out to obtain a curve-optimized animation file;

[0056] Adjust the local gravity of the curve optimization animation file to obtain the gravity change animation file;

[0057] Enhance the inertia of the gravity change animation file to obtain a physical stylized animation file;

[0058] Enhance the details of the physical stylized animation file to obtain the final artistic animation file.

[0059] In a second aspect, the present application further provides a motion capture and evaluation device, comprising:

[0060] Source module, used to obtain motion capture data;

[0061] The repair module is used to repair the motion capture data and obtain standardized motion data;

[0062] Animation module, used to obtain virtual character animation files based on standardized action data and virtual character skeleton model;

[0063] Evaluation module, used to evaluate virtual character animation files in multiple dimensions and generate a comprehensive evaluation report;

[0064] The adjustment module is used to adjust the virtual character animation file based on the comprehensive evaluation report to obtain the final animation file.

[0065] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method provided in the first aspect of the present application are implemented:

[0066] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method provided in the first aspect of the present application.

[0067] The aforementioned motion capture and evaluation method, apparatus, device, and medium acquire motion capture data; repair the motion capture data to obtain standardized motion data; generate a virtual character animation file based on the standardized motion data and a virtual character skeletal model; perform a multi-dimensional evaluation of the virtual character animation file to generate a comprehensive evaluation report; and adjust the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file. Through these steps, motion capture animations are evaluated, improving animation quality and increasing work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0069] Figure 1 A flow chart of a motion capture and evaluation method of the present invention;

[0070] Figure 2 A diagram of a motion capture and evaluation device according to the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0072] In one embodiment, Figure 1 As shown, a motion capture and evaluation method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, or to a system including a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0073] Step 101: Acquire motion capture data.

[0074] Specifically, motion capture data is raw motion data generated by capturing the motion trajectory of a real person or object using sensors, optical devices, or inertial measurement units. Motion capture equipment is used to capture the movements of real actors or objects and record their motion information.

[0075] For example, the motion capture data may include joint positions, rotations, accelerations, etc.

[0076] Step 102: Repair the motion capture data to obtain standardized motion data.

[0077] Specifically, standardized motion data is in a unified data format. The repair operation can make the data complete and coherent, meeting the input requirements of the virtual character skeleton model, while eliminating errors caused by equipment or environment.

[0078] Exemplary repairs may include:

[0079] (1) Denoising: smoothing jittery data through filtering algorithms; (2) Frame filling: using interpolation algorithms to supplement missing frames; (3) Format unification: converting data into a standard coordinate system, fixed frame rate, etc.

[0080] Step 103: Obtain a virtual character animation file based on the standardized motion data and the virtual character skeleton model.

[0081] Specifically, the skeletal model of an avatar is a hierarchical structure composed of joints and bones that defines the character's motion logic. Animation files contain skeletal motion data. By mapping standardized motion data to the corresponding joints of the avatar's skeleton and adjusting the motion scale and range of motion, we can initially generate animations that match the character model.

[0082] Step 104: perform a multi-dimensional evaluation on the virtual character animation file and generate a comprehensive evaluation report.

[0083] Specifically, multi-dimensional evaluations provide quantitative or qualitative analysis of animation quality from perspectives such as fluidity, physical plausibility, and artistic expression. A comprehensive evaluation report includes a list of issues, scoring metrics, and optimization suggestions. This evaluation identifies animation flaws by examining issues such as clipping, leg slippage, and joint limit violations.

[0084] Step 105: Adjust the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file.

[0085] Specifically, based on the evaluation report, animation parameters are modified, clipping issues are fixed manually or through algorithms, motion curves are adjusted to enhance the natural feel, and the beginning and end frames of the looping animation are ensured to be seamless. The final animation file is a finished animation that meets technical requirements and artistic standards.

[0086] This embodiment provides a motion capture and evaluation method, which evaluates the motion capture animation through the steps of acquiring motion capture data; repairing the motion capture data to obtain standardized motion data; obtaining a virtual character animation file based on the standardized motion data and a virtual character skeleton model; performing a multi-dimensional evaluation of the virtual character animation file to generate a comprehensive evaluation report; and adjusting the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file, thereby improving animation quality and work efficiency.

[0087] In one embodiment, a multi-dimensional evaluation of a virtual character animation file is performed to generate a comprehensive evaluation report, including:

[0088] Step 201 : Based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal coordination analysis; the coordination evaluation report is used to characterize the temporal synchronization and spatial coordination efficiency of different body parts of the character in the virtual character animation file.

[0089] Specifically, the Coordination Assessment Report is a document used to quantify the synchronization and collaborative efficiency of character movements. Algorithms can be used to detect whether the start, peak, and end times of each body part's movements match, calculate the consistency of movement directions, detect conflicting movements, and mark inconsistent frame intervals. Spatiotemporal coordination is the coordinated movement relationship between different body parts of a character across time and space. Spatiotemporal coordination is the coordinated movement relationship between different body parts of a character across time and space.

[0090] Step 202 : obtaining a fluency evaluation report based on the motion trajectory in the virtual character animation file; the fluency evaluation report is used to characterize the action continuity of the motion trajectory.

[0091] Specifically, a fluency assessment report is a document describing the continuity of an action, including the locations of frames experiencing jitter, sudden changes, or speed discontinuities. This report can be generated by analyzing the first and second derivatives of the motion trajectory to detect sudden changes, compare the ideal motion template with the actual trajectory, quantify the degree of deviation, and mark any jittery frames.

[0092] Step 203 , based on the joint angles of the virtual character animation file, extract the out-of-limit joint frames and generate a rationality evaluation report; the rationality evaluation is used to characterize the biomechanical rationality of the action.

[0093] Specifically, the plausibility assessment report is a document that identifies out-of-limit joint positions, angles, and durations. Joint ranges are defined based on the character type. Joint Euler angles or quaternion data are traversed frame by frame, and out-of-limit angles are screened. The problem type, which can include elbow hyperextension and spinal twisting, is then classified. Finally, a plausibility assessment report is generated. Out-of-limit joint frames are abnormal frames where joint rotation angles exceed the preset biomechanical range.

[0094] Step 204 : Based on the virtual character animation file and the virtual character physical parameters, verify the center of mass projection and generate a physical inertia evaluation report; the physical inertia evaluation report is used to characterize the physical credibility of the action.

[0095] Specifically, the Physical Inertia Assessment Report verifies that the center of mass motion complies with physical laws. Based on the character's physical parameters, including mass distribution and bone length, the center of mass position is calculated in real time. The center of mass projection is checked to see if it lies within the support surface. The relationship between motion acceleration and character mass is compared, detecting abnormal inertia and momentum, and generating a Physical Inertia Assessment Report. The center of mass projection is the vertical projection of the character's center of mass onto the support surface.

[0096] Step 205 : Integrate the coordination assessment report, the fluency assessment report, the rationality assessment report, and the physical inertia assessment report to generate a comprehensive assessment report.

[0097] Specifically, the comprehensive assessment report prioritizes coordination, fluidity, rationality, and physical inertia. The scores for each dimension are weighted and summarized, graded by severity, and output as a visual chart.

[0098] This embodiment transforms subjective animation evaluation into objective data indicators through multi-dimensional analysis, accurately locates problems and generates executable optimization solutions, significantly improving the realism, artistic expression and technical compliance of actions.

[0099] In one embodiment, based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal coordination analysis, including:

[0100] Step 301: Based on the upper and lower limb movements of the virtual character animation file, dynamic time warping is used to calculate the minimum path cumulative cost using the following formula to obtain the optimal alignment path:

[0101]

[0102] Among them, C(i,j) is the minimum path cumulative cost from the starting point to the current point, a i is the data of the i-th frame of the reference sequence, b j is the data of the jth frame of the target sequence, d(d i ,b j ) is the local distance of the quantity sequence in frame (i, j).

[0103] Specifically, the optimal alignment path is the path that minimizes the cumulative cost of aligning two sequences. This can be done by inputting upper and lower limb motion sequences, calculating the difference between each pair of frames using Euclidean distance or cosine similarity, and then backfilling the matrix using a recursive formula to find the path with the minimum cost. Dynamic Time Warping is a nonlinear matching algorithm that aligns two time series (such as upper and lower limb motions) to measure similarity.

[0104] In step 302, based on the optimal alignment path, the average action delay is obtained using the following formula:

[0105]

[0106] Where a is the average action delay, N is the total length of the optimal alignment path, i k ,j k are the frame numbers of the reference sequence and target sequence corresponding to the kth step of the path.

[0107] Specifically, the reference frame and target frame of each step are extracted from the optimal alignment path, and the result is calculated by the formula. A positive value indicates that the lower limb lags behind the upper limb, and a negative value indicates that it is ahead. The global timing deviation index is obtained to preliminarily judge the direction of movement dislocation.

[0108] Step 303: Based on the average action delay, a timing deviation exceeding limit frame is obtained; the timing deviation exceeding limit frame indicates that the upper limb and lower limb movements are out of sync.

[0109] Specifically, a timing deviation exceeding limit frames is a frame interval in which the difference between the corresponding reference and target sequences in the alignment path exceeds a preset threshold. Based on project requirements, we define the maximum allowable deviation, mark all frames that meet the threshold, and generate a list of exceeding frames and deviation values. This allows us to pinpoint specific action segments with significant misalignment and prioritize key problem areas for repair.

[0110] Step 304 : extract the dominant motion in the virtual character animation file through principal component analysis to obtain the motion contribution.

[0111] Specifically, the motion contribution is the variance contribution of each principal component, reflecting its weighted influence on the overall movement. Principal component analysis (PCA) normalizes the joint motion data, decomposes data correlations, extracts eigenvectors and eigenvalues, and retains principal components with a total contribution greater than 85% to determine the dominant motion chain. The dominant motion is the principal component that characterizes the main direction of motion variation, as determined by PCA decomposition.

[0112] Step 305 : Compare the motion contribution with the biomechanical standard to obtain the abnormal dominant motion chain; the biomechanical standard is extracted from the biomechanical reference standard library.

[0113] Specifically, abnormal dominant kinematic chains are principal components whose actual contributions significantly deviate from the standard. Based on the movement type, a reference contribution range can be retrieved from a library, and principal components that exceed the allowable deviation can be marked. This generates a list of abnormal kinematic chains and identifies compensatory behaviors that violate biomechanics. The biomechanical standard is a predefined standard kinematic chain contribution.

[0114] Step 306: Determine a coordination assessment report based on the timing deviation exceeding limit frames and the abnormal dominant motion chain.

[0115] Specifically, the coordination assessment report is a comprehensive analysis document that integrates timing deviations and abnormal motion chains. It can include problematic frames and contribution deviations. By associating timing violation frames with abnormal motion chains, grading the problem severity, and outputting a visual chart, the coordination assessment report is generated.

[0116] This embodiment uses timing alignment and motion decomposition, combined with a biomechanical standard library, to transform subjective "motion incoordination" into objective spatiotemporal deviation parameters and contribution anomaly indicators, accurately locating the root cause of the problem, providing a data-driven solution for animation optimization, and significantly improving the naturalness and professionalism of virtual character movements.

[0117] In one embodiment, a fluency evaluation report is obtained based on the motion trajectory in the virtual character animation file, including:

[0118] Step 401 : Based on the virtual character animation file, obtain the joint angular acceleration by using the second-order difference method.

[0119] Specifically, the joint angle sequence is extracted from the animation file, and the second-order difference formula is applied to each frame to obtain the angular acceleration sequence. The acceleration values ​​that match the original data timeline are retained to quantify the intensity of the dynamic changes in joint motion. The second-order difference method calculates the second-order derivative from discrete data points and can reveal transient sudden changes in movement that do not conform to biomechanics, such as abnormal elbow tremor when swinging an arm.

[0120] Step 402 : Compare the joint angular acceleration with a threshold value to obtain an acceleration exceeding limit frame.

[0121] Specifically, acceleration-exceeding frames are frames where the absolute value of the joint angular acceleration exceeds a preset threshold. This can be used to troubleshoot non-physical motion caused by data acquisition or binding errors, as well as stuttering and twitching during positioning. Based on the joint type, thresholds are loaded from the biomechanics database. All frames where the absolute value of acceleration exceeds the threshold are marked, and their times and acceleration values ​​are recorded to generate a list of exceeding-the-limit frames.

[0122] Step 403: Fit the hand trajectory in the virtual character animation file based on the Bezier curve, and obtain the curvature standard deviation using the following formula:

[0123] B(t)=(x(t),y(t))

[0124]

[0125] Among them, σ κ is the standard deviation of curvature, B(T) is the Bezier curve parameter equation of the hand trajectory, x(t), y(t) are the horizontal and vertical coordinates of the hand at time t, κ(t) is the curvature of the hand trajectory at time t, x′(t), y′(t) are the first-order derivatives of the hand trajectory at time t, x”(t), y”(t) are the second-order derivatives of the hand trajectory at time t, N is the total number of frames of the trajectory, κ i is the curvature value of the i-th frame, μ κ is the average value of the curvature.

[0126] Specifically, the curvature standard deviation measures the fluctuation in the curvature of the trajectory; larger values ​​indicate more dramatic changes in curvature. The hand position sequence is obtained from the animation file. The least squares method is used to determine the control points of the Bezier curve, allowing the curve to approximate the original trajectory. Based on the parametric equation of the fitted curve, the curvature is calculated frame by frame, and the mean curvature of all frames is calculated, followed by the standard deviation. Bezier curve fitting is used to generate a smooth curve approximating the original hand trajectory using the control points.

[0127] Step 404 : Compare the curvature standard deviation with a threshold to obtain an unnatural linear motion frame.

[0128] Specifically, unnatural linear motion frames are defined as frames where the curvature standard deviation falls below a lower threshold or exceeds an upper threshold. These frames are used to identify trajectories that violate natural motion patterns and ensure that hand movements conform to human kinematics. A curvature fluctuation range is set based on the motion type. Frames below the range are labeled as mechanical linear motion, while frames above the range are labeled as jittery trajectories. The problem type is categorized by frame interval, and unnatural linear motion frames are generated.

[0129] Step 405 : Obtain a fluency evaluation report based on the acceleration exceeding limit frames and the unnatural linear motion frames.

[0130] Specifically, the two types of problem frames are correlated and a visual chart is output to create a smoothness assessment report. This report integrates frames with excessive acceleration and frames with unnatural linear motion, and includes the problem type, location, and optimization suggestions. By using quantitative indicators (acceleration, curvature) and problem location, it can systematically improve animation smoothness and reduce the cost of manual frame-by-frame inspection.

[0131] This embodiment combines dynamic and kinematic analysis to convert subjective "unsmoothness" into quantifiable problems of excessive acceleration and abnormal curvature, accurately locates defects such as jitter, lag, and mechanical movement, provides animators with a data-driven correction basis, and significantly improves the naturalness and realism of virtual character movements.

[0132] In one embodiment, repairing the motion capture data to obtain standardized motion data includes:

[0133] Step 501: De-noise the motion capture data to obtain a de-noised motion data file.

[0134] Specifically, filtering algorithms can be used to eliminate high-frequency noise, jitter, and instantaneous mutations in motion capture data, thereby preserving the true motion trend.

[0135] Step 502 : performing singular value decomposition based on the denoised motion data file to obtain a rotation matrix and a translation vector from the denoised motion data file to the virtual world coordinate system.

[0136] Specifically, the denoised static pose is selected as the reference frame, and singular value decomposition (SVD) is performed on the key points of the motion capture device coordinate system and the virtual world coordinate system to obtain the coordinate system transformation relationship. SVD is a mathematical method used to decompose a data matrix into a rotation matrix, a translation vector, and a scaling factor.

[0137] Step 503 : transforming the joint data in the denoised motion data file into a unified virtual world coordinate system based on the rotation matrix and the translation vector to obtain a motion data file with aligned coordinate systems.

[0138] Specifically, the virtual world coordinate system is a predefined global coordinate system. The original coordinates of each joint are converted into virtual coordinates, and matrix operations are performed on the joint positions and rotations of all frames to generate a data file of coordinate system one.

[0139] Step 504 : Reconstruct the motion data file based on the PCA of the coordinate system alignment, repair the missing data, and obtain a complete motion data file.

[0140] Specifically, principal components are calculated using complete frame data, and missing joint data is approximated through linear combinations of the existing principal components to generate a complete motion file. PCA reconstruction is used to fill in missing frames based on principal component analysis and leverage motion data correlations, thus repairing short-term data loss and maintaining motion coherence.

[0141] Step 505: Band-pass filter the complete motion data file to obtain a respiratory layer signal; the respiratory layer signal is an independent chest cavity rise and fall signal.

[0142] Specifically, a bandpass frequency corresponding to the respiratory rhythm is set, and chest joint motion data is filtered to generate a respiratory layer signal file. The respiratory layer signal is a low-frequency movement related to the rise and fall of the chest cavity, independent of limb movement. This can separate breathing and limb movement, preserving the amplitude and rhythm of breathing.

[0143] Step 506 , removing the respiratory layer signal from the complete motion data file to obtain the main motion layer signal; the main motion layer signal is pure limb motion data.

[0144] Specifically, the main motion layer signal removes non-breathing body movement data. Subtracting the breathing layer signal from the complete data, the positions of other joints are adjusted to maintain posture balance, generating a pure body motion file. This allows for the separation of artistic control (breathing) from technical movement (swinging the sword), enhancing production flexibility.

[0145] Step 507: resample and align the respiratory layer signal and the main motion layer signal to obtain standardized motion data.

[0146] Specifically, resampling alignment is used to unify the sampling rate and timeline of the breathing layer and the main motion layer, which can eliminate phase delay caused by filtering and ensure natural synchronization between breathing and movement. Linear interpolation is performed on the breathing layer signal to match the frame rate and duration of the main motion layer, ensuring synchronization of breathing and movement during key events.

[0147] This implementation uses denoising, coordinate alignment, PCA reconstruction, respiration separation, and resampling to restore raw motion capture data into highly accurate, complete, and standardized motion data that meets the requirements of virtual production. This addresses core issues such as multi-device compatibility, data loss, and biological motion separation, significantly reducing animation production costs and improving quality. Preserving the respiration layer signal allows for more flexible motion processing and evaluation.

[0148] In one embodiment, a virtual character animation file is obtained based on the standardized motion data and the virtual character skeleton model, including:

[0149] Step 601: Align the standard human skeleton structure with the character mesh model to obtain a skeleton-bound character model.

[0150] Specifically, a rigged character model is a 3D model that associates standard bones with the character mesh's vertices. A standard bone structure is a predefined, universal bone hierarchy, including joint names, positions, and parent-child relationships. Aligning the key points of the standard bones with the corresponding positions on the character model, adjusting the bone lengths to suit the character's body shape, and generating a rigged model file.

[0151] Step 602 , calculate the vertex weights of the character model bound to the skeleton using a heat diffusion algorithm to obtain a skeleton model of the virtual character; the vertex weights affect the degree of model deformation when the joints are bent.

[0152] Specifically, vertex weights represent how strongly a mesh vertex is influenced by a particular bone. The heat diffusion algorithm simulates the process of heat diffusion from a bone to the mesh. It calculates the weight of the bone's influence on the vertex, sets the bone as the heat source, and calculates the distribution of vertex weights using partial differential equations to ensure that the sum of each vertex weight is 1. This vertex weight is then used to adjust the bones accordingly.

[0153] Step 603: Adjust the motion amplitude of the standardized motion data according to the virtual character skeleton model to obtain an initial calibration animation file.

[0154] Specifically, the skeletal length ratios of the motion-captured actor and the avatar are compared, and the displacements and rotations in the motion data are linearly or nonlinearly scaled to generate a pre-calibrated animation file tailored to the character's body type. Motion range calibration standardizes the motion data based on the avatar's skeletal proportions, preventing motion range discrepancies and providing a preliminary solution to range of motion conflicts.

[0155] Step 604 : Based on the initial calibration animation file and the virtual character skeleton model, the end effector position is calculated by inverse kinematics to obtain a preliminary redirection animation file.

[0156] Specifically, inverse kinematics (IK) is used to reversely calculate the rotation angles of each joint in the skeletal chain based on the target position of the end effector. The preliminary redirected animation file is the animation file after the end position is corrected by IK. The end position in the motion capture data is set as the inverse kinematics target, and the gradient descent or CCD algorithm is used to iteratively solve the joint angle so that the end approaches the target. An animation file with an accurate match of the end position is generated to ensure the accuracy of the contact points of the character's hands and feet and correct the end offset caused by differences in bone proportions.

[0157] Step 605: Based on the physical parameters of the virtual character, physical constraints are applied to the preliminary redirected animation file to obtain a constrained animation file.

[0158] Specifically, physical constraints are rigid restrictions based on the character's mass and joint range of motion. Constraint parameters are read from the character's configuration file, and joint angles that violate the constraints are clamped or interpolated to smooth them. This generates a physically correct post-constraint animation file, eliminating anti-joint and out-of-range motion, and enhancing the biomechanical plausibility of the movements.

[0159] Step 606 : Based on the physical parameters of the virtual character, balance the center of mass of the constrained animation file to obtain a physically corrected animation file.

[0160] Specifically, center of mass balance adjusts the character's posture so that its projection always lies within the supporting polygon. The center of mass position is calculated in real time based on the character's mass distribution. An optimization algorithm fine-tunes the upper body's posture to maintain balance, generating physically corrected animation files to enhance the dynamic rationality of the movements.

[0161] Step 607: Smooth the key frame curves of the physically corrected animation file to obtain the virtual character animation file.

[0162] Specifically, the Euler angles or quaternion curves of the joints are derived from the animation file, and cubic spline interpolation or Bezier curve fitting is applied to reduce acceleration abrupt changes, resulting in a smooth final animation file. Keyframe curve smoothing is used to interpolate and optimize animation curves (such as the change in joint rotation angle over time), eliminating abrupt changes.

[0163] This embodiment accurately adapts standardized motion data to the skeleton model of any virtual character, solving core problems such as body shape differences and stiff movements, significantly reducing manual correction costs and improving cross-project reusability.

[0164] In one embodiment, after adjusting the virtual character animation file based on the comprehensive evaluation report to obtain the final animation file, the method further includes:

[0165] Step 701: Receive a preset art enhancement instruction; the art enhancement instruction is used to instruct stylization of the final animation file.

[0166] Specifically, artistic enhancement instructions are user-defined stylized parameters that guide the artistic expression of animation. By converting natural language instructions into technical parameters, matching style templates from a preset library, and generating executable stylized configuration parameter tables, the system can transform subjective artistic requirements into programmable animation adjustment rules.

[0167] Step 702 : Based on the art enhancement instruction, adjust the key frames of the final animation file to obtain a preliminary stylized animation file.

[0168] Specifically, keyframe adjustment involves modifying the displacement, rotation, or time position of a keyframe, changing the amplitude or rhythm of the action. By following instructions, keyframe displacements can be enlarged or reduced, and keyframe intervals can be stretched or compressed to generate a preliminary stylized animation file, achieving exaggerated or refined movements.

[0169] Step 703: Adjust the initial stylized animation file to ease in and out to obtain a curve-optimized animation file.

[0170] Specifically, an ease-in is when the acceleration gradually increases from zero at the start of an action, and an ease-out is when the acceleration gradually decreases to zero at the end of an action. Ease-ins and ease-outs are used to adjust the slope of the animation curve and control the acceleration changes at the start and end of an action, eliminating a mechanical feel and enhancing the tension of the action.

[0171] Step 704: adjust the local gravity of the curve optimization animation file to obtain a gravity change animation file.

[0172] Specifically, local gravity adjustment involves modifying the gravity coefficient for a specific body part or time period. By binding independent gravity parameters to specific bones or model components within the animation engine and controlling the gravity field's effective range through keyframes, you can create surreal motion effects and enhance the atmosphere of a scene.

[0173] Step 705: Inertia enhancement is performed on the gravity change animation file to obtain a physical stylized animation file.

[0174] Specifically, by simulating the inertia of movement, the impact or duration of the movement is increased, the intensity of the movement is strengthened, and the realism of the details is increased.

[0175] For example, when turning quickly, delay the upper body movement; add speed-based follow-up animation for accessories (such as long hair and scarf).

[0176] Step 706: Enhance the details of the physical stylized animation file to obtain a final artistic animation file.

[0177] Specifically, secondary movements such as breathing fluctuations, muscle contractions, and changes in clothing folds are injected to adjust facial details.

[0178] This embodiment transforms basic animation into a final work that conforms to a specific artistic style through instruction parsing and detail enhancement, quantifying parameter control, physical simulation, and detail carving to balance technical accuracy and artistic expression.

[0179] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0180] Based on the same inventive concept, the present application also provides a motion capture and assessment device for implementing the aforementioned motion capture and assessment method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more of the following embodiments of a motion capture and assessment device can be found in the above-mentioned limitations of a motion capture and assessment method, and will not be further elaborated here.

[0181] In an exemplary embodiment, Figure 2 As shown, a motion capture and evaluation device 800 is provided, comprising:

[0182] Source module 801, used to obtain motion capture data;

[0183] A repair module 802 is used to repair the motion capture data to obtain standardized motion data;

[0184] Animation module 803, for obtaining a virtual character animation file based on the standardized motion data and the virtual character skeleton model;

[0185] Evaluation module 804, for performing multi-dimensional evaluation on the virtual character animation file and generating a comprehensive evaluation report;

[0186] The adjustment module 805 is used to adjust the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file.

[0187] Furthermore, the evaluation module 804 is further configured to:

[0188] Based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal coordination analysis. The coordination evaluation report is used to characterize the temporal synchronization and spatial coordination efficiency of different body parts of the virtual character animation file.

[0189] Based on the motion trajectory in the virtual character animation file, a fluency evaluation report is obtained; the fluency evaluation report is used to characterize the action continuity of the motion trajectory;

[0190] Based on the joint angles of the virtual character animation file, the over-limit joint frames are extracted and a rationality assessment report is generated; the rationality assessment is used to characterize the biomechanical rationality of the action;

[0191] Based on the virtual character animation file and the virtual character's physical parameters, the center of mass projection is verified and a physical inertia evaluation report is generated; the physical inertia evaluation report is used to characterize the physical credibility of the action;

[0192] Integrate the coordination assessment report, fluency assessment report, rationality assessment report and physical inertia assessment report to generate a comprehensive assessment report.

[0193] Furthermore, the evaluation module 804 is further configured to:

[0194] Based on the upper and lower limb movements of the avatar animation file, dynamic time warping is used to calculate the minimum path cumulative cost using the following formula to obtain the optimal alignment path:

[0195]

[0196] Among them, C(i,j) is the minimum path cumulative cost from the starting point to the current point, a i is the data of the i-th frame of the reference sequence, b j is the data of the jth frame of the target sequence, d(a i ,b j ) is the local distance of the quantity sequence in frame (i, j);

[0197] Based on the optimal alignment path, the average action delay is obtained by the following formula:

[0198]

[0199] Where a is the average action delay, N is the total length of the optimal alignment path, i k ,j k is the frame number of the reference sequence and target sequence corresponding to the kth step of the path;

[0200] Based on the average action delay, the timing deviation exceeding the limit frame is obtained; the timing deviation exceeding the limit frame indicates that the upper and lower limb movements are out of sync;

[0201] The dominant motion in the virtual character animation file is extracted through principal component analysis to obtain the motion contribution;

[0202] The abnormal dominant kinematic chain is obtained by comparing the movement contribution with the biomechanical standard, which is extracted from the biomechanical reference standard library.

[0203] Based on the timing deviation exceeding the limit frame and the abnormal dominant motion chain, the coordination assessment report is determined.

[0204] Furthermore, the evaluation module 804 is further configured to:

[0205] Based on the virtual character animation file, the joint angular acceleration is obtained by the second-order difference method:

[0206] Compare the joint angular acceleration with the threshold to obtain the acceleration exceeding limit frame;

[0207] Based on the Bezier curve fitting of the hand trajectory in the virtual character animation file, the curvature standard deviation is obtained using the following formula:

[0208] B(t)=(x(t),y(t))

[0209]

[0210] Among them, σ k is the standard deviation of curvature, B(T) is the Bezier curve parameter equation of the hand trajectory, x(t), y(t) are the horizontal and vertical coordinates of the hand at time t, κ(t) is the curvature of the hand trajectory at time t, x′(t), y′(t) are the first-order derivatives of the hand trajectory at time t, x”(t), y”(t) are the second-order derivatives of the hand trajectory at time t, N is the total number of frames of the trajectory, κ i is the curvature value of the i-th frame, μ k is the average value of curvature;

[0211] Compare the curvature standard deviation to the threshold to obtain the unnatural linear motion frame;

[0212] A smoothness evaluation report is obtained based on acceleration exceeding limit frames and unnatural linear motion frames.

[0213] Furthermore, the repair module 802 is further configured to:

[0214] De-noise the motion capture data to obtain the denoised motion data file;

[0215] Perform singular value decomposition on the denoised motion data file to obtain the rotation matrix and translation vector of the denoised motion data file to the virtual world coordinate system;

[0216] The joint data in the denoised motion data file is converted to a unified virtual world coordinate system based on the rotation matrix and translation vector to obtain a coordinate-aligned motion data file;

[0217] PCA reconstruction of the motion data file based on coordinate system alignment repairs missing data to obtain a complete motion data file;

[0218] The complete motion data file is band-pass filtered to obtain the respiratory layer signal; the respiratory layer signal is an independent chest cavity rise and fall signal;

[0219] Remove the respiratory layer signal from the complete motion data file to obtain the main motion layer signal; the main motion layer signal is pure limb movement data;

[0220] The respiratory layer signal and the main motion layer signal are resampled and aligned to obtain standardized motion data.

[0221] Furthermore, the animation module 803 is further configured to:

[0222] Align the standard human skeleton structure with the character mesh model to obtain a skeleton-bound character model;

[0223] The vertex weights of the character model with bound bones are calculated using a heat diffusion algorithm to obtain the skeleton model of the virtual character; the vertex weights affect the degree of model deformation when the joints are bent;

[0224] Adjust the standardized motion data movement amplitude according to the virtual character skeleton model to obtain the initial calibration animation file;

[0225] Based on the initial calibration animation file and the virtual character skeleton model, the end effector position is solved by inverse kinematics to obtain the preliminary redirection animation file;

[0226] Based on the physical parameters of the virtual character, physical constraints are applied to the preliminary redirected animation file to obtain the constrained animation file;

[0227] Based on the virtual character's physical parameters, the center of mass of the constrained animation file is balanced to obtain a physically corrected animation file;

[0228] The key frame curve of the physically corrected animation file is smoothed to obtain the virtual character animation file.

[0229] Furthermore, the system also includes an artistic module for:

[0230] receiving preset art enhancement instructions; the art enhancement instructions are used to instruct stylization of the final animation file;

[0231] Based on the art enhancement instructions, adjust the key frames of the final animation file to obtain a preliminary stylized animation file;

[0232] Adjust the initial stylized animation file in and out to obtain a curve-optimized animation file;

[0233] Adjust the local gravity of the curve optimization animation file to obtain the gravity change animation file;

[0234] Enhance the inertia of the gravity change animation file to obtain a physical stylized animation file;

[0235] Enhance the details of the physical stylized animation file to obtain the final artistic animation file.

[0236] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the motion capture and evaluation method as described above when executing the computer program.

[0237] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0238] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0239] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A motion capture and evaluation method, characterized in that: The method comprises: Get motion capture data; Repairing the motion capture data to obtain standardized motion data; Obtaining a virtual character animation file based on the standardized motion data and the virtual character skeleton model; Evaluate the virtual character animation file in multiple dimensions and generate a comprehensive evaluation report; The virtual character animation file is adjusted based on the comprehensive evaluation report to obtain a final animation file.

2. A motion capture and evaluation method according to claim 1, characterized in that: The multi-dimensional evaluation of the virtual character animation file to generate a comprehensive evaluation report includes: Based on the virtual character animation file, a coordination evaluation report is obtained through spatiotemporal coordination analysis; the coordination evaluation report is used to characterize the temporal synchronization and spatial coordination efficiency of different body parts of the character in the virtual character animation file; Obtaining a fluency evaluation report based on the motion trajectory in the virtual character animation file; wherein the fluency evaluation report is used to characterize the action continuity of the motion trajectory; Based on the joint angles of the virtual character animation file, extracting the out-of-limit joint frames and generating a rationality evaluation report; the rationality evaluation is used to characterize the biomechanical rationality of the action; Based on the virtual character animation file and the virtual character physical parameters, verify the center of mass projection and generate a physical inertia evaluation report; the physical inertia evaluation report is used to characterize the physical credibility of the action; The coordination assessment report, the fluency assessment report, the rationality assessment report and the physical inertia assessment report are integrated to generate the comprehensive assessment report.

3. A motion capture and evaluation method according to claim 2, characterized in that: The coordination evaluation report is obtained by analyzing the spatiotemporal coordination of the virtual character animation file, including: Based on the upper and lower limb movements of the virtual character animation file, dynamic time warping is used to calculate the minimum path cumulative cost using the following formula to obtain the optimal alignment path: Among them, C(i,j) is the minimum path cumulative cost from the starting point to the current point, a i is the data of the i-th frame of the reference sequence, b j is the data of the jth frame of the target sequence, d(a i ,b j ) is the local distance of the quantity sequence in frame (i, j); Based on the optimal alignment path, the average action delay is obtained by the following formula: Where a is the average action delay, N is the total length of the optimal alignment path, i k ,j k is the frame number of the reference sequence and target sequence corresponding to the kth step of the path; Based on the average action delay, a timing deviation exceeding limit frame is obtained; the timing deviation exceeding limit frame indicates that the upper limb and lower limb movements are out of sync; extracting the dominant motion in the virtual character animation file through principal component analysis to obtain motion contribution; Comparing the movement contribution with a biomechanical standard to obtain an abnormal dominant movement chain; the biomechanical standard is extracted from a biomechanical reference standard library; The coordination assessment report is determined based on the timing deviation exceeding limit frame and the abnormal dominant motion chain.

4. A motion capture and evaluation method according to claim 2, characterized in that: The obtaining of a fluency evaluation report based on the motion trajectory in the virtual character animation file includes: Based on the virtual character animation file, obtaining the joint angular acceleration by a second-order difference method; Comparing the joint angular acceleration with a threshold value to obtain an acceleration exceeding limit frame; Based on the Bezier curve fitting of the hand trajectory in the virtual character animation file, the curvature standard deviation is obtained by the following formula: B(t)=(x(t),y(t)) Among them, σ κ is the standard deviation of curvature, B(T) is the Bezier curve parameter equation of the hand trajectory, x(t), y(t) are the horizontal and vertical coordinates of the hand at time t, κ(t) is the curvature of the hand trajectory at time t, x′(t), y′(t) are the first-order derivatives of the hand trajectory at time t, x”(t), y”(t) are the second-order derivatives of the hand trajectory at time t, N is the total number of frames of the trajectory, κ i is the curvature value of the i-th frame, μ κ is the average value of curvature; Comparing the curvature standard deviation with a threshold value to obtain an unnatural linear motion frame; A smoothness evaluation report is obtained based on the acceleration exceeding limit frames and the unnatural linear motion frames.

5. A motion capture and evaluation method according to claim 1, characterized in that: The step of repairing the motion capture data to obtain standardized motion data includes: Denoising the motion capture data to obtain a denoised motion data file; Performing singular value decomposition based on the denoised motion data file to obtain a rotation matrix and a translation vector from the denoised motion data file to a virtual world coordinate system; Converting the joint data in the denoised motion data file to a unified virtual world coordinate system based on the rotation matrix and the translation vector to obtain a coordinate-aligned motion data file; PCA reconstruction of the motion data file aligned with the coordinate system is performed to repair missing data and obtain a complete motion data file; Band-pass filtering is performed on the complete motion data file to obtain a respiratory layer signal; the respiratory layer signal is an independent chest cavity rise and fall signal; The complete motion data file is stripped of the respiratory layer signal to obtain a main motion layer signal; the main motion layer signal is pure limb motion data; The respiratory layer signal and the main motion layer signal are resampled and aligned to obtain the standardized motion data.

6. A motion capture and evaluation method according to any one of claims 1 to 5, characterized in that: The step of obtaining a virtual character animation file based on the standardized motion data and the virtual character skeleton model includes: Align the standard human skeleton structure with the character mesh model to obtain a skeleton-bound character model; Calculating the vertex weights of the character model bound to the skeleton using a heat diffusion algorithm to obtain the skeleton model of the virtual character; the vertex weights affect the degree of model deformation when the joints are bent; Adjusting the movement amplitude of the standardized movement data according to the virtual character skeleton model to obtain an initial calibration animation file; Based on the initial calibration animation file and the virtual character skeleton model, the end effector position is calculated by inverse kinematics to obtain a preliminary redirection animation file; Based on the physical parameters of the virtual character, physically constraining the preliminary redirected animation file to obtain a constrained animation file; Based on the physical parameters of the virtual character, balancing the center of mass of the constrained animation file to obtain a physically corrected animation file; The key frame curve of the physically corrected animation file is smoothed to obtain the virtual character animation file.

7. A motion capture and evaluation method according to claim 1, characterized in that: After adjusting the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file, the method further includes: receiving a preset art enhancement instruction; the art enhancement instruction is used to instruct stylization of the final animation file; Based on the art enhancement instruction, adjusting the key frames of the final animation file to obtain a preliminary stylized animation file; Adjusting the preliminary stylized animation file in and out to obtain a curve-optimized animation file; Adjusting the local gravity of the curve optimization animation file to obtain a gravity change animation file; Inertia enhancement of the gravity change animation file to obtain a physical stylized animation file; The details of the physical stylized animation file are enhanced to obtain a final artistic animation file.

8. A motion capture and evaluation device, characterized in that: The device comprises: Source module, used to obtain motion capture data; A repair module, used for repairing the motion capture data to obtain standardized motion data; Animation module, used for obtaining a virtual character animation file based on the standardized motion data and the virtual character skeleton model; An evaluation module, configured to evaluate the virtual character animation file in multiple dimensions and generate a comprehensive evaluation report; An adjustment module is used to adjust the virtual character animation file based on the comprehensive evaluation report to obtain a final animation file.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images