Virtual fitting fit assessment method based on dynamic posture

Abstract

The present application relates to the technical field of virtual fitting, and particularly relates to a virtual fitting fit evaluation method based on dynamic posture, comprising the following steps: acquiring dynamic posture data of a user when the user performs a preset action sequence, constructing a digital fabric physical model containing physical parameters of tensile stiffness, bending stiffness and shear stiffness, and calculating the tensile, wrinkle and fit changes of clothes under the driving of the dynamic posture data to generate dynamic deformation data; subsequently, calculating the initial gap distribution and initial pressure mapping between the clothes and the body based on the dynamic deformation data, obtaining a stable gap distribution and an accurate pressure mapping through space-time consistency analysis; finally, generating fit evaluation indexes containing a dynamic fit index, a local compression coefficient and an overall fit score, and outputting a fit report through visual rendering. The present application can reflect the fit state of clothes under dynamic movement, and improve the fine evaluation capability of virtual fitting.

Description

Technical Field

[0001] This invention relates to the field of virtual fitting technology, and in particular to a method for assessing the fit of virtual fitting based on dynamic posture. Background Technology

[0002] With the development of virtual reality, 3D reconstruction, and computer graphics technologies, virtual try-on technology has gradually become an important tool for apparel e-commerce, clothing customization, and online size selection. Through 3D human body modeling and fabric simulation, users can view the appearance and general fit of clothing without actually wearing it. However, traditional virtual try-on relies heavily on static postures and single-frame human body models, making it difficult to reflect the dynamic fit of clothing caused by changes in movement during actual wearing scenarios, such as shoulder displacement during walking, underarm compression when raising the arm, or lateral waist tension during twisting. Furthermore, the physical properties of clothing materials, such as stretching, bending, and shearing, significantly affect the wrinkles, tightness, and compression of clothing during dynamic movement. Therefore, more refined dynamic simulation technology is needed to realistically represent the wearing effect.

[0003] Existing virtual fitting technologies often suffer from several drawbacks, including an inability to accurately capture user dynamic postures, simulate fabric deformation under complex movements, and provide quantifiable fit metrics. On one hand, current technologies typically use simplified skeletal models or pose estimation based solely on depth maps, making it difficult to obtain high-precision body joint positions and motion trajectories, resulting in inaccurate clothing deformation simulations. On the other hand, existing fabric simulation algorithms often rely on static collision detection and coarse physical approximations, failing to maintain energy conservation under rapid movements and unable to prevent clothing from penetrating the human body model. Furthermore, existing solutions often lack quantitative analysis of the gap distribution and pressure mapping between clothing and the human body, failing to provide evaluation metrics such as dynamic fit index and local pressure coefficient, and unable to automatically generate structured fit reports and visualizations based on evaluation results. This makes it difficult for users to intuitively understand the fit issues and limits the ability to optimize clothing design and personalize adjustments. Summary of the Invention

[0004] This invention provides a virtual fitting fit assessment method based on dynamic posture. By acquiring dynamic posture data, simulating dynamic deformation of clothing, calculating gap and pressure distribution, and generating a visual fit report, it achieves a precise quantitative assessment of the dynamic fit performance of clothing.

[0005] A virtual fitting fit assessment method based on dynamic posture includes the following steps:

[0006] S1: Acquire the user's dynamic posture data, which includes the user's body joint positions and movement trajectories when performing a preset action sequence;

[0007] S2: Based on the dynamic posture data, simulate the dynamic behavior of the clothing on the user and generate dynamic deformation data of the clothing, wherein the dynamic behavior includes the stretching, wrinkling and fitting changes of the clothing.

[0008] S3: Based on the dynamic deformation data of the clothing, assess the fit of the clothing and generate fit assessment indicators, which include the gap distribution and pressure mapping between the clothing and the body.

[0009] S4: Output the fit assessment index to generate a visualized fit report, which includes fit scores and improvement suggestions under dynamic postures.

[0010] Optionally, S1 includes:

[0011] S11: Obtain a preset action sequence, which includes multiple standard action instructions to guide the user to perform dynamic postures. The preset action sequence is provided to the user through user interface display or voice prompts, and a signal confirming the start of the action is received from the user to generate an executable preset action sequence.

[0012] S12: Based on the preset action sequence, capture multi-frame motion data of the user performing the preset action sequence in real time using a depth camera or inertial measurement unit sensor. The multi-frame motion data includes three-dimensional point cloud information of the user's body in the time series or acceleration and angular velocity data, and output the multi-frame motion data.

[0013] S13: Calculate the user's body joint position and motion trajectory from the multi-frame motion data, wherein the body joint position is extracted from the multi-frame motion data by a skeletal tracking algorithm or machine learning model, and the motion trajectory is obtained by analyzing the changes in the body joint position over time, and output the body joint position and motion trajectory.

[0014] S14: Generate the dynamic posture data based on the body joint positions and motion trajectories, wherein the dynamic posture data includes a body joint position matrix and a motion trajectory vector, which are used as input for subsequent steps.

[0015] Optionally, generating an executable preset action sequence includes: selecting and combining a predefined action library to form a personalized preset action sequence based on the user's basic body data or historical fitting records, wherein the basic body data includes height, weight, and body type classification.

[0016] Optionally, the dynamic posture data further includes: performing data standardization processing on the body joint positions and motion trajectories, removing abnormal frame data during the execution of the preset action sequence, and supplementing missing body joint positions to form complete and consistent dynamic posture data.

[0017] Optionally, S2 includes:

[0018] S21: Construct a digital fabric physical model of the garment, wherein the digital fabric physical model includes physical parameters of tensile stiffness, bending stiffness and shear stiffness of the fabric, and generate a digital fabric physical model based on the physical parameters.

[0019] S22: Couple the digital fabric physical model with the dynamic posture data obtained from step S1, and calculate the real-time deformation of the digital fabric physical model under the action of the dynamic posture data through the physics engine to generate preliminary deformation data of the garment, wherein the real-time deformation includes the stretching, wrinkling and fitting changes of the garment.

[0020] S23: Perform dynamic physical effect correction on the preliminary deformation data of the clothing. Based on the principle of energy conservation and the fabric collision detection algorithm, correct the non-penetrating constraint relationship between the clothing and the body, and output the corrected dynamic deformation data of the clothing. The dynamic deformation data includes precise stretching distribution, wrinkle depth and fit change.

[0021] Optionally, the construction of the digital fabric physical model of the garment includes: acquiring physical sample data of the target fabric through a fabric mechanics testing instrument, and inputting the physical sample data into a material parameter inversion algorithm to calibrate the tensile stiffness, bending stiffness and shear stiffness physical parameters in the digital fabric physical model.

[0022] Optionally, S3 includes:

[0023] S31: Based on the dynamic deformation data of the clothing obtained from step S2, calculate the signed distance field between the clothing mesh model and the user body mesh model, generate the initial gap distribution, and at the same time calculate the initial pressure mapping through a predefined pressure-stress transformation function based on the stress tensor in the dynamic deformation data of the clothing.

[0024] S32: Perform spatiotemporal consistency analysis on the initial gap distribution and initial pressure mapping, smooth inter-frame abrupt changes through time series filters, and fill in the data missing due to occlusion through spatial interpolation algorithms to form a stable gap distribution and accurate pressure mapping.

[0025] S33: Based on the stable gap distribution and accurate pressure mapping, the dynamic fit index, local pressure coefficient and overall fit score are calculated and integrated to generate the final fit evaluation index, wherein the fit evaluation index includes the quantified gap distribution value and pressure mapping value.

[0026] Optionally, the calculation of the signed distance field between the clothing mesh model and the user body mesh model includes: establishing a spatial distance query data structure based on the user body mesh model, traversing the shortest directed distance from each vertex of the clothing mesh model to the user body mesh model, where positive values ​​represent gaps and negative values ​​represent penetrations, thereby generating the initial gap distribution.

[0027] Optionally, S4 includes:

[0028] S41: Parse the obtained fit assessment indicators, and process the dynamic fit index, local pressure coefficient, overall fit score, gap distribution value and pressure mapping value in the fit assessment indicators into a structured form according to a predefined report template to generate structured fit data.

[0029] S42: Based on the structured fit data, automatically generate a draft fit report containing a comprehensive fit score and specific improvement suggestions. The comprehensive fit score is obtained by weighted fusion calculation of various values ​​in the structured fit data. The specific improvement suggestions are generated by matching the abnormal gap distribution values ​​and abnormal pressure mapping values ​​in the structured fit data with the pre-stored clothing alteration knowledge base through a rule engine.

[0030] S43: Visualize and render the initial draft of the fit report, mark the overall fit score in the form of stars, and overlay and fuse the specific improvement suggestions with keyframe images of the user's dynamic posture, and output the final visualized fit report through the augmented reality interface.

[0031] Optionally, the step of structuring the fit assessment indicators according to a predefined report template includes: verifying the data integrity and filtering outliers for the gap distribution values ​​and pressure mapping values ​​in the fit assessment indicators, and storing the verified data in the corresponding data structure according to body region partitions to generate complete and standardized structured fit data.

[0032] The beneficial effects of this invention are:

[0033] This invention, by acquiring dynamic posture data including body joint position matrices and motion trajectory vectors, can accurately characterize the user's three-dimensional posture changes under different action states. Compared to traditional fitting methods that rely solely on static postures or simplified skeletal models, this invention can dynamically capture the user's true body posture during complex movements such as walking, raising arms, and rotating. Combined with a constructed digital fabric physical model containing physical parameters of tensile stiffness, bending stiffness, and shear stiffness, it can realistically simulate the stretching, wrinkling, and fit changes of clothing under dynamic driving conditions. Furthermore, by using the principle of energy conservation and a fabric collision detection algorithm to correct clothing deformation, the dynamic deformation data more closely matches the physical behavior of real fabric, thereby significantly improving the accuracy and reliability of fit assessment.

[0034] This invention calculates a signed distance field using a spatial distance query data structure to obtain an initial gap distribution that distinguishes between gaps and penetrations. It then uses a pressure-stress transformation function to generate an initial pressure mapping, achieving a true quantification of the distance and pressure relationship between clothing and the user's body. Subsequently, a Kalman filter is used to smooth abrupt changes in the time series, and radial basis function interpolation is employed to fill in data gaps caused by occlusion, ensuring the gap distribution and pressure mapping have continuity, stability, and integrity in both time and space. Through these multi-level spatiotemporal consistency processing methods, this invention can provide high-precision, reliable gap and pressure data for key areas (such as shoulders, chest, waist, and thighs), providing a solid data foundation for dynamic fit index, local pressure coefficient, and overall fit rating.

[0035] This invention structures gap distribution values, pressure mapping values, dynamic fit index, local compression coefficient, and overall fit score according to a predefined template. A weighted fusion calculation is then used to obtain a comprehensive fit score. A rule engine automatically retrieves pattern adjustment solutions from a clothing alteration knowledge base to generate specific improvement suggestions. This ensures that the final fit report not only includes quantitative indicators but also provides clear optimization directions. Furthermore, using WebGL technology and shader programs, gap distribution and pressure mapping are overlaid and rendered as heatmaps onto the user's body model and fused with dynamic pose keyframe images. The final output is displayed through an augmented reality interface, allowing users to intuitively view the location and trend of fit issues. This invention offers significant advantages in user experience and clothing adjustment guidance and can be widely applied in scenarios such as virtual try-on, e-commerce size selection, and clothing customization. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in this 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 for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the S1 process in an embodiment of the present invention. Detailed Implementation

[0039] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0040] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0041] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0042] like Figures 1-2 As shown, a virtual fitting fit assessment method based on dynamic posture includes the following steps:

[0043] S1: Obtain the user's dynamic posture data, which includes the user's body joint positions and movement trajectories when performing a preset action sequence, specifically:

[0044] S11: Displays multiple standard action commands to the user through a pre-defined human-computer interaction interface. These commands are displayed graphically one by one on the interface, while a voice prompt module outputs a description of each command. First, the system reads the user's basic body data, including height, weight, and body type, and then extracts previously performed actions from stored historical fitting records.

[0045] Based on body shape data and historical fitting records, and following preset action matching rules, multiple standard action instructions are selected from a predefined action library and combined in a predetermined order to form a personalized preset action sequence.

[0046] Once a personalized preset action sequence is formed, all the actions in the preset action sequence are listed on the interface, and voice prompts are played simultaneously to inform the user of the order and requirements of the actions to be performed.

[0047] When a user confirms the start of execution via the confirmation button on the interface or a voice confirmation command, the personalized preset action sequence is determined as an executable preset action sequence.

[0048] Subsequently, the executable preset action sequences are recorded in chronological order to guide the acquisition of subsequent multi-frame motion data.

[0049] S12: Based on the executable preset action sequence, activate the depth camera and at least one inertial measurement unit (IMU) sensor to begin acquiring multi-frame motion data. The depth camera continuously acquires data at a preset frame rate throughout the entire process of the user performing the preset action sequence, obtaining 3D point cloud information for each time frame. The IMU sensor is worn on a designated part of the user's body and acquires acceleration and angular velocity data at the same time intervals as the depth camera.

[0050] During the data acquisition process, to ensure the consistency of multi-frame motion data, the 3D point cloud information acquired by the depth camera and the acceleration and angular velocity data acquired by the inertial measurement unit (IMU) sensor are strictly aligned according to timestamps. Using a pre-defined sensor fusion algorithm, the aligned 3D point cloud information, acceleration data, and angular velocity data are calibrated and integrated. A coordinate transformation step is performed to bring the coordinate systems of different sensors to the same reference coordinate system, and weighted fusion is then performed based on the noise characteristics of the sensors.

[0051] Through the above steps, the fused motion information at each time frame is obtained, and the fusion results of all time frames are arranged in chronological order to form multi-frame motion data.

[0052] The multi-frame motion data contains user motion information for all time periods corresponding to the preset action sequence, which is used to calculate the body joint position and motion trajectory in step S13.

[0053] S13: Read the multi-frame motion data obtained in S12 and calculate the user's body joint positions from the multi-frame motion data using a machine learning model.

[0054] Human body structure is identified using a pre-trained machine learning model. The machine learning model is built using a pose estimation algorithm based on convolutional neural networks. It extracts and matches features from multiple frames of motion data to output the position of body joints at each time frame.

[0055] After obtaining the positions of the body joints in chronological order, the motion trajectory of each body joint is calculated based on the changes in the body joint positions over time.

[0056] Specifically, differential analysis and interpolation are performed on the body joint positions of adjacent time frames to obtain the motion path that changes over time, thus forming a continuous motion trajectory. Through the above steps, the body joint positions and motion trajectories corresponding to the execution process of the preset action sequence are obtained.

[0057] After step S13 is completed, the body joint positions and motion trajectories are output as input for step S14 to generate dynamic posture data.

[0058] S14: Generate dynamic posture data based on the body joint positions and motion trajectories output in step S13.

[0059] First, the body joint positions and motion trajectories are standardized. The standardization process includes normalizing the body joint positions in all time frames according to a unified reference coordinate system, and scaling the motion trajectory according to a preset range, so that the body joint positions and motion trajectories of different users can be represented on a uniform scale.

[0060] After data standardization, abnormal frame data generated during the execution of the preset action sequence is detected. Abnormal frame data includes frames with severe displacement of body joint positions due to acquisition errors and frames with discontinuous changes in motion trajectory over time. Abnormal frame data is identified using preset threshold rules and then removed from subsequent processing.

[0061] After removing abnormal frame data, missing body joint positions caused by abnormal frame removal, short-term occlusion, or sensor frame loss are filled in. During the filling process, the missing body joint positions at the time points are calculated using interpolation methods based on the body joint positions of adjacent time frames, thereby ensuring the temporal continuity of body joint positions throughout the entire preset action sequence.

[0062] After standardization, anomaly removal, and missing data completion, the body joint positions across all time frames are arranged chronologically and by joint number to construct a body joint position matrix. The motion trajectories of all body joints across each time frame are then concatenated in a preset order to construct motion trajectory vectors. The body joint position matrix and motion trajectory vectors together constitute the dynamic pose data.

[0063] Finally, the dynamic posture data, which includes the body joint position matrix and motion trajectory vector, is used as the input for step S2, and is then used for the subsequent dynamic behavior simulation and dynamic deformation calculation of the clothing.

[0064] S2: Based on dynamic posture data, simulate the dynamic behavior of clothing on the user, and generate dynamic deformation data of the clothing. The dynamic behavior includes the stretching, wrinkling, and fit changes of the clothing, specifically:

[0065] S21: First, a digital fabric physical model of the garment is constructed. During the construction process, to ensure that the digital fabric physical model can accurately reflect the real physical properties of the target fabric, physical sample data of the target fabric is obtained through a fabric mechanics testing instrument. The fabric mechanics testing instrument applies tensile load, bending load, and shear load to the target fabric according to a preset loading method, and records the mechanical response of the fabric under tension, bending, and shear forces respectively.

[0066] The aforementioned physical sample data is input into the material parameter inversion algorithm. Based on the mechanical response characteristics of the target fabric, the material parameter inversion algorithm calibrates the tensile stiffness, bending stiffness, and shear stiffness in the model. By minimizing the error between the sample data and the model's predicted values, the algorithm ensures that the physical parameters of tensile stiffness, bending stiffness, and shear stiffness in the digital fabric physical model are consistent with those of the real fabric.

[0067] After completing the material parameter inversion, a digital fabric physical model is constructed based on the calibrated tensile stiffness, bending stiffness and shear stiffness physical parameters, and this digital fabric physical model is used for the dynamic coupling calculation in step S22.

[0068] The inversion objective function is expressed as:

[0069] ;

[0070] in, This represents the objective function value of the inversion algorithm, used to measure the difference between the simulated force-displacement curve and the actual test curve. The number of data points participating in the inversion calculation. Digital fabric physical model in the first The simulated force values ​​calculated at each displacement point Fabric mechanical testing instrument at the The actual force value measured at each displacement point.

[0071] S22: Couple the digital fabric physical model constructed in step S21 with the dynamic posture data obtained in step S1. The dynamic posture data includes body joint positions and motion trajectories. By using body joint positions as driving boundaries and motion trajectories as driving forces, the digital fabric physical model undergoes real-time deformation under the influence of dynamic posture.

[0072] To achieve real-time deformation calculation, a position dynamics-based physics engine is used to iteratively solve the digital fabric physical model. In each time frame, the position dynamics-based physics engine determines the driving conditions of the clothing mesh based on the body joint positions and motion trajectories, and iteratively updates each vertex in the digital fabric physical model frame by frame. During the iteration process, the mechanical response of the fabric elements is calculated based on the tensile stiffness, bending stiffness, and shear stiffness physical parameters in the digital fabric physical model. By continuously solving for the positional changes of each vertex, the clothing undergoes stretching, wrinkling, and conformation changes under dynamic posture driving.

[0073] Iterative updates are represented as:

[0074] ;

[0075] in, This represents the 3D position of a vertex in the clothing mesh at the current timestamp. This represents the updated position of the vertex at the next timestamp. The time step for iterative solving is consistent with the sampling frequency of the dynamic attitude data. This is the resultant acceleration experienced by the vertex at the current timestamp, including accelerations caused by tensile force, bending force, shear force, and dynamic inertial force.

[0076] After the iterative solution converges, the preliminary deformation data of the clothing is obtained. The preliminary deformation data is the dynamic response result of the clothing under dynamic posture, including the stretching, wrinkling and fitting changes of the clothing, and is used as the input data for step S23.

[0077] S23: To improve the physical accuracy of clothing deformation, dynamic physical effect correction is performed on the preliminary deformation data of clothing obtained in step S22.

[0078] First, based on the principle of energy conservation, the kinetic energy and elastic potential energy generated by the dynamic posture in the initial deformation data of the clothing are calculated. Based on the distribution relationship between kinetic and elastic potential energy, the amplitude of the initial deformation of the clothing is adjusted to conform to the energy distribution law of real fabric under rapid movement.

[0079] After energy correction, a cloth collision detection algorithm is used to detect penetration between clothing vertices and the body mesh model reconstructed from dynamic pose data. In penetration detection, the spatial distance between the clothing vertices and the body mesh model is calculated to determine if mesh overlap has occurred. When penetration is detected, according to the non-penetration constraint rules of the cloth collision detection algorithm, the penetrating clothing vertex is moved along the normal direction to a non-penetrating position outside the body mesh model surface.

[0080] After collision correction, the dynamic deformation data of the corrected clothing is obtained. The dynamic deformation data includes the corrected precise stretch distribution, wrinkle depth, and fit change, and serves as the input for the subsequent step S3.

[0081] S3: Based on the dynamic deformation data of the clothing, assess the fit of the clothing and generate fit assessment indicators. The fit assessment indicators include the distribution of gaps and pressure mapping between the clothing and the body, specifically:

[0082] S31: First, read the dynamic deformation data output in step S2, which includes the stretching distribution, wrinkle depth and fit change of the clothing under dynamic posture.

[0083] After acquiring the dynamic deformation data, a clothing mesh model is established based on the dynamic deformation data of the clothing, and a user body mesh model is reconstructed based on the dynamic posture data generated in step S1.

[0084] To calculate the signed distance field between the clothing mesh model and the user body mesh model, a spatial distance query data structure is established based on the user body mesh model.

[0085] The spatial distance query data structure adopts a three-dimensional spatial acceleration structure based on the user's body mesh model, which is used to efficiently query the shortest directed distance from any vertex in the clothing mesh model to the surface of the user's body mesh model.

[0086] During the distance calculation process, each vertex of the clothing mesh model is traversed, and the shortest directed distance from that vertex to the surface of the user's body mesh model is calculated. A positive value of this shortest directed distance indicates that there is a gap between the clothing and the body, while a negative value indicates that the clothing vertex penetrates the user's body surface. By calculating the shortest directed distances to all clothing mesh vertices, an initial gap distribution between the clothing and the body is generated.

[0087] After calculating the initial gap distribution, the initial pressure mapping is calculated using a predefined pressure-stress transformation function based on the stress tensor in the dynamic deformation data.

[0088] The predefined pressure-stress conversion function maps the tensile and shear stresses contained in the stress tensor to the corresponding force applied per unit area to the user's skin.

[0089] An initial pressure mapping is generated by performing point-by-point transformation on the stress tensor of each vertex of the clothing mesh model.

[0090] The predefined pressure-stress transformation function is expressed as:

[0091] ;

[0092] in, In the initial pressure mapping, the clothing mesh model of the first... Each vertex corresponds to a force per unit area acting on human skin. For the stress tensor in the dynamic deformation data, in the clothing mesh model... Stress tensor at each vertex In the first The unit normal vector at each vertex, determined by the surface normal direction of the user's body mesh model, is used to characterize the direction of pressure application.

[0093] S32: Perform spatiotemporal consistency analysis on the initial gap distribution and initial pressure mapping obtained in step S31 to improve the stability and integrity of the gap distribution and pressure mapping.

[0094] First, a time-series filter is applied to smooth the initial gap distribution and initial pressure mapping across consecutive time frames. The time-series filter uses a Kalman filter, which optimally weights the estimated and predicted values ​​for each time frame to suppress random noise in the data, smooth abrupt changes between frames, and ensure the continuity and consistency of the gap distribution and pressure mapping in the time dimension.

[0095] After completing the temporal smoothing process, a spatial interpolation algorithm is used to fill in the data gaps caused by occlusion. The spatial interpolation algorithm adopts the radial basis function interpolation method. By using the vertices of the clothing mesh in the unoccluded area of ​​the dynamic deformation data as known points, a radial basis function model with the spatial position of the vertices as parameters is constructed, and the gap data and pressure data of the occluded area are reconstructed by interpolation.

[0096] This interpolation method enables the shaded area to obtain consistent and coherent gap and pressure values ​​with the surrounding area, thereby forming a complete and stable gap distribution and accurate pressure mapping.

[0097] S33: Based on the stable gap distribution and accurate pressure mapping obtained in step S32, calculate the final fit evaluation index.

[0098] The fit assessment indicators include dynamic fit index, local pressure coefficient and overall fit score, and include quantified gap distribution value and pressure mapping value.

[0099] When calculating the dynamic fit index, based on a stable gap distribution and accurate pressure mapping, the calculation is performed on the entire time series covered by the preset key body areas over the dynamic posture data.

[0100] For each key body region, the interval uniformity index and pressure comfort index of that region are calculated over the entire time series, and the two are weighted and averaged with preset weights.

[0101] The final weighted average is normalized, and the normalized result is used as the dynamic fit index.

[0102] The weighted average of key body regions is expressed as follows:

[0103] ;

[0104] in, The weighted average index is calculated based on a stable gap distribution and accurate pressure mapping within a pre-defined key body area. The median value of the dynamic fit index before normalization is also included. This refers to the gap uniformity index calculated over the entire time series covered by dynamic attitude data in this key body region. This refers to the pressure comfort index calculated over the entire time series covered by dynamic posture data for this key body region. This is a weighting coefficient for the gap uniformity index, used to characterize the importance of gap distribution in the calculation of the dynamic fit index. This is a weighting coefficient for the pressure comfort index, used to characterize the importance of pressure comfort in the calculation of the dynamic fit index.

[0105] After obtaining the dynamic fit index, it is combined with the local compression coefficient and overall fit score to form the final fit assessment index. This fit assessment index is used to describe the overall fit performance of the garment under dynamic postures, providing basic data for the subsequent generation of visualization reports.

[0106] S4: Based on the fit assessment metrics, generate a visualized fit report, including fit scores under dynamic postures and improvement suggestions, specifically:

[0107] S41: First, analyze the fit assessment indicators obtained in step S3. These indicators include the dynamic fit index, local pressure coefficient, overall fit score, gap distribution value, and pressure mapping value. During analysis, each indicator is formatted according to a predefined report template, ensuring that all indicators are organized according to a unified data type and field structure.

[0108] Subsequently, based on the predefined report template, data integrity verification was performed on the gap distribution values ​​and pressure mapping values ​​in the fit assessment indicators.

[0109] Data integrity verification includes detecting whether data is missing, whether there are frames inconsistent with the dynamic pose sequence, and whether there are abnormal data points that exceed the numerical range.

[0110] Outlier values ​​detected are filtered out by using threshold rules to remove gap distribution values ​​and pressure mapping values ​​that do not meet the data stability requirements.

[0111] After data validation and filtering are completed, the validated gap distribution values ​​and pressure mapping values ​​are divided according to predefined body regions, and the values ​​in each region are stored in the corresponding data structure.

[0112] This regionalized storage method generates complete and standardized structured fit data, providing a unified data foundation for further analysis in step S42.

[0113] S42: Based on the structured fit data obtained in step S41, automatically generate a draft fit report containing a comprehensive fit score and specific improvement suggestions.

[0114] First, a comprehensive fit score is obtained by weighted fusion calculation of various values ​​in the structured fit data. The weighted fusion calculation is based on preset fusion weights and comprehensively processes the dynamic fit index, local pressure coefficient, overall fit score, regional gap distribution value, and pressure mapping value, so that the comprehensive fit score can objectively reflect the overall fit performance of the clothing under dynamic posture.

[0115] The weighted fusion of the overall fit score is expressed as follows:

[0116] ;

[0117] in, The overall fit score is obtained by weighted fusion of structured fit data. The number of indicators involved in the fusion calculation includes dynamic fit index, local pressure coefficient, overall fit score, gap distribution value, pressure mapping value, etc. No. Numerical values ​​in structured, fit-to-the-fit data, such as gap distribution values ​​or pressure mapping values ​​for a certain region. In order to be with the first The weights corresponding to each indicator are derived from the preset weight configuration.

[0118] Subsequently, the rule engine is invoked to perform matching analysis on the abnormal gap distribution values ​​and abnormal pressure mapping values ​​in the structured fit data.

[0119] The rules engine matches the values ​​of each body region in the structured fit data with a pre-stored clothing alteration knowledge base. When it detects that the gap distribution value of a specific body region in the structured fit data continuously exceeds a preset threshold, it automatically retrieves the pattern adjustment scheme corresponding to that body region from the clothing alteration knowledge base and incorporates the pattern adjustment scheme as part of the specific improvement suggestions into the initial draft of the fit report.

[0120] Through the above steps, a draft fit report containing a comprehensive fit score and specific improvement suggestions is generated, providing a content basis for the visualization rendering in step S43.

[0121] S43: Visualize and render the initial draft of the fit report generated in step S42. During the rendering process, the overall fit score is displayed in the form of stars. The score range is converted into the corresponding number of star icons through preset visual mapping rules, so that the overall fit score is presented in an intuitive form.

[0122] Subsequently, the specific improvement suggestions were overlaid and fused with keyframe images of the user's dynamic pose. The keyframe images were selected from the dynamic pose data, reflecting the main actions or key poses, and were used as the background layer.

[0123] To achieve visual fusion between layers, WebGL technology is used to load the overall fit score, specific improvement suggestions, and keyframe images as independent layers, and then combine and render these layers through a shader program.

[0124] The shader program generates a corresponding heatmap visual texture based on the gap distribution value and pressure mapping value. The heatmap texture is then overlaid and rendered on the user's body model in the form of a color gradient, allowing the user to intuitively view the fit of different areas of the body.

[0125] Ultimately, a visual fit report is output through the augmented reality interface, allowing users to view the overall fit score, specific improvement suggestions, and the visualization effect of dynamic pose keyframes and heatmaps overlaid in the augmented reality environment.

[0126] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0127] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for assessing the fit of virtual clothing based on dynamic posture, characterized in that, Includes the following steps: Acquire the user's dynamic posture data, which includes the user's body joint positions and movement trajectories when performing a preset sequence of actions; Based on the dynamic posture data, the dynamic behavior of the clothing on the user is simulated to generate dynamic deformation data of the clothing, wherein the dynamic behavior includes the stretching, wrinkling and fitting changes of the clothing. Based on the dynamic deformation data of the clothing, the fit of the clothing is evaluated, and a fit evaluation index is generated. The fit evaluation index includes the gap distribution and pressure mapping between the clothing and the body. Based on the aforementioned fit assessment metrics, a visual fit report is generated, which includes fit scores and improvement suggestions under dynamic postures. The generated fit assessment indicators include: Based on dynamic deformation data, the signed distance field between the clothing mesh model and the user body mesh model is calculated to generate the initial gap distribution. At the same time, based on the stress tensor in the dynamic deformation data of the clothing, the initial pressure mapping is calculated through a predefined pressure-stress transformation function. Spatiotemporal consistency analysis is performed on the initial gap distribution and initial pressure mapping. Inter-frame abrupt changes are smoothed by time series filters, and data loss caused by occlusion is filled by spatial interpolation algorithms to form a stable gap distribution and accurate pressure mapping. Based on the stable gap distribution and accurate pressure mapping, the dynamic fit index, local pressure coefficient and overall fit score are calculated and integrated to generate the final fit evaluation index, wherein the fit evaluation index includes quantified gap distribution values ​​and pressure mapping values. The calculation of the signed distance field between the clothing mesh model and the user body mesh model includes: establishing a spatial distance query data structure based on the user body mesh model, traversing the shortest directed distance from each vertex of the clothing mesh model to the user body mesh model, where positive values ​​represent gaps and negative values ​​represent penetrations, thereby generating the initial gap distribution.

2. The virtual fitting fit assessment method based on dynamic posture according to claim 1, characterized in that, The acquisition of the user's dynamic posture data includes: Obtain a preset action sequence, which includes multiple standard action instructions to guide the user to perform dynamic postures. The preset action sequence is provided to the user through user interface display or voice prompts, and a signal confirming the start by the user is received to generate an executable preset action sequence. Based on the preset action sequence, multi-frame motion data of the user performing the preset action sequence is captured in real time by a depth camera or inertial measurement unit sensor. The multi-frame motion data includes three-dimensional point cloud information of the user's body in the time series or acceleration and angular velocity data, and multi-frame motion data is output. The user's body joint positions and motion trajectories are calculated from the multi-frame motion data. The body joint positions are extracted from the multi-frame motion data through a machine learning model, and the motion trajectories are obtained by analyzing the changes in body joint positions over time. The body joint positions and motion trajectories are then output. The dynamic posture data is generated based on the body joint positions and motion trajectories, wherein the dynamic posture data includes a body joint position matrix and a motion trajectory vector, which are used as input for subsequent steps.

3. The virtual fitting fit assessment method based on dynamic posture according to claim 2, characterized in that, The generation of executable preset action sequences includes: selecting and combining personalized preset action sequences from a predefined action library based on the user's basic body data or historical fitting records, wherein the basic body data includes height, weight, and body type classification.

4. The virtual fitting fit assessment method based on dynamic posture according to claim 2, characterized in that, The dynamic posture data also includes: performing data standardization processing on the body joint positions and movement trajectories, removing abnormal frame data during the execution of the preset action sequence, and filling in missing body joint positions to form complete and consistent dynamic posture data.

5. The virtual fitting fit assessment method based on dynamic posture according to claim 2, characterized in that, The dynamic deformation data of the generated clothing includes: A digital fabric physical model of clothing is constructed, which includes physical parameters of tensile stiffness, bending stiffness and shear stiffness of the fabric, and a digital fabric physical model is generated based on the physical parameters. The digital fabric physical model is coupled with dynamic posture data, and the real-time deformation of the digital fabric physical model under the action of the dynamic posture data is calculated by the physics engine to generate preliminary deformation data of the garment. The real-time deformation includes the stretching, wrinkling and fitting changes of the garment. The initial deformation data of the garment is corrected by dynamic physical effects. Based on the principle of energy conservation and fabric collision detection algorithm, the non-penetrating constraint relationship between the garment and the body is corrected, and the corrected dynamic deformation data of the garment is output. The dynamic deformation data includes precise stretching distribution, wrinkle depth and fit change.

6. The virtual fitting fit assessment method based on dynamic posture according to claim 5, characterized in that, The construction of the digital fabric physical model for clothing includes: acquiring physical sample data of the target fabric through a fabric mechanics testing instrument, and inputting the physical sample data into a material parameter inversion algorithm to calibrate the tensile stiffness, bending stiffness, and shear stiffness physical parameters in the digital fabric physical model.

7. The virtual fitting fit assessment method based on dynamic posture according to claim 1, characterized in that, The generated visualization of the fit report includes: The obtained fit assessment indicators are analyzed, and the dynamic fit index, local pressure coefficient, overall fit score, gap distribution value and pressure mapping value in the fit assessment indicators are structured according to the predefined report template to generate structured fit data. Based on the structured fit data, a draft fit report containing a comprehensive fit score and specific improvement suggestions is automatically generated. The comprehensive fit score is obtained by weighted fusion calculation of various values ​​in the structured fit data. The specific improvement suggestions are generated by a rule engine matching the abnormal gap distribution values ​​and abnormal pressure mapping values ​​in the structured fit data with a pre-stored clothing alteration knowledge base. The initial draft of the fit report is visualized and rendered, the overall fit score is marked in the form of stars, and the specific improvement suggestions are overlaid and fused with keyframe images of the user's dynamic posture. The final visualized fit report is then output through an augmented reality interface.

8. The virtual fitting fit assessment method based on dynamic posture according to claim 7, characterized in that, The process of structuring the fit assessment indicators according to a predefined report template includes: verifying the data integrity and filtering outliers for the gap distribution values ​​and pressure mapping values ​​in the fit assessment indicators, and storing the verified data in the corresponding data structure according to body region partitions to generate complete and standardized structured fit data.

Images