Face aging detection method, system, electronic device and computer program product

Abstract

The application is suitable for the field of computer image processing, and provides a face aging detection method and system, an electronic device and a computer program product, wherein the method comprises: acquiring a first three-dimensional mesh model and a second three-dimensional mesh model of a target face; aligning the first three-dimensional mesh model and the second three-dimensional mesh model to the same space coordinate system based on a plurality of set facial feature points, and establishing a vertex mapping relationship corresponding to one-to-one between a plurality of first vertices and a plurality of second vertices; calculating a projection difference value of each first vertex relative to the corresponding second vertex; determining the sagging information and the drooping information of the plurality of first vertices according to the plurality of projection difference values, and correspondingly generating and outputting an aging detection report of the target face. The scheme can determine the sagging information and the drooping information of the target face based on the three-dimensional mesh model of the target face, correspondingly generate the aging detection report of the target face, and realize face aging detection based on sagging features and drooping features.

Description

Technical Field

[0001] This application belongs to the field of computer image processing, and in particular relates to a method, system, electronic device and computer program product for detecting facial aging. Background Technology

[0002] In the medical aesthetics industry, the degree of facial aging is crucial information for developing effective cosmetic treatment plans. Current mainstream methods primarily rely on two-dimensional image analysis technology to determine the degree of facial aging by capturing skin surface features (such as roughness, pigmentation area, and pore density). There is currently a lack of facial aging detection methods based on three-dimensional structural features such as depressions (caused by volume loss) and sagging (caused by fascia layer relaxation). Summary of the Invention

[0003] This application provides a method, system, electronic device, and computer program product for detecting facial aging, in order to solve the problem that the prior art does not combine concave and drooping features to achieve facial aging detection.

[0004] The first aspect of this application provides a method for detecting facial aging, including:

[0005] Obtain a first 3D mesh model and a second 3D mesh model of the target face; the first 3D mesh model contains multiple first vertices, the second 3D mesh model contains multiple second vertices, and the first 3D mesh model and the second 3D mesh model also contain multiple defined facial feature points;

[0006] Based on multiple defined facial feature points, the first three-dimensional mesh model and the second three-dimensional mesh model are aligned to the same spatial coordinate system, and a one-to-one vertex mapping relationship is established between multiple first vertices and multiple second vertices accordingly.

[0007] Calculate the projection difference of each first vertex relative to its corresponding second vertex;

[0008] Based on the multiple projection differences, the concavity information and drooping information of the multiple first vertices are determined;

[0009] Based on the concavity and drooping information of multiple first vertices, an aging detection report of the target face is generated and output.

[0010] A second aspect of this application provides a facial aging detection system, comprising:

[0011] The acquisition module is used to acquire a first three-dimensional mesh model and a second three-dimensional mesh model of the target face; the first three-dimensional mesh model contains multiple first vertices, the second three-dimensional mesh model contains multiple second vertices, and the first three-dimensional mesh model and the second three-dimensional mesh model also contain multiple set facial feature points;

[0012] The alignment mapping module is used to align the first three-dimensional mesh model and the second three-dimensional mesh model to the same spatial coordinate system based on multiple set facial feature points, and to establish a one-to-one vertex mapping relationship between multiple first vertices and multiple second vertices accordingly.

[0013] The calculation module is used to calculate the projection difference of each first vertex relative to its corresponding second vertex;

[0014] The determining module is used to determine the concavity information and drooping information of multiple first vertices based on multiple projection differences;

[0015] The output generation module is used to generate and output an aging detection report of the target face based on the concavity and drooping information of multiple first vertices.

[0016] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described in the first aspect.

[0017] A fourth aspect of this application provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0018] A fifth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0019] As can be seen from the above, this application aligns the first and second 3D mesh models with the same spatial coordinate system based on the predefined facial feature points contained in both the first and second 3D mesh models. A one-to-one vertex mapping relationship is then established between multiple first vertices of the first 3D mesh model and multiple second vertices of the second 3D mesh model within this spatial coordinate system. This allows for the calculation of the projection difference between each first vertex and its corresponding second vertex, quantifying the concavity and drooping information of each first vertex. This, in turn, generates and outputs an aging detection report for the target face, providing a face aging detection method based on concavity and drooping features. This fills the gap in traditional methods for 3D structural aging detection and solves the problem in existing technologies that do not combine concavity and drooping features to achieve face aging detection. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a facial aging detection method provided in an embodiment of this application;

[0022] Figure 2 This is a schematic diagram of a three-dimensional chromatographic model of aging distribution provided in an embodiment of this application;

[0023] Figure 3 This is a structural diagram of a face aging detection system provided in an embodiment of this application;

[0024] Figure 4 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0026] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0027] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0028] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0029] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [the described condition or event] is detected," or "in response to detection of [the described condition or event]."

[0030] In specific implementations, the terminals described in the embodiments of this application include, but are not limited to, other portable devices such as mobile phones, laptop computers, or tablet computers with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads). It should also be understood that in some embodiments, the device is not a portable communication device, but a desktop computer with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads).

[0031] The following discussion describes terminals that include displays and touch-sensitive surfaces. However, it should be understood that terminals may include one or more other physical user interface devices such as physical keyboards, mice, and / or joysticks.

[0032] The terminal supports a variety of applications, such as one or more of the following: drawing applications, presentation applications, word processing applications, website creation applications, disc burning applications, spreadsheet applications, game applications, telephone applications, video conferencing applications, email applications, instant messaging applications, exercise support applications, photo management applications, digital camera applications, digital camcorder applications, web browsing applications, digital music player applications, and / or digital video player applications.

[0033] Various applications that can run on a terminal can use at least one common physical user interface device, such as a touch-sensitive surface. One or more functions of the touch-sensitive surface and the corresponding information displayed on the terminal can be adjusted and / or changed between and / or within applications. In this way, the terminal's common physical architecture (e.g., the touch-sensitive surface) can support various applications with user interfaces that are intuitive and transparent to the user.

[0034] It should be understood that the sequence number of each step in this embodiment does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this application embodiment.

[0035] To illustrate the technical solution described in this application, specific embodiments are provided below.

[0036] See Figure 1 , Figure 1 This is a flowchart of a facial aging detection method provided in an embodiment of this application. Figure 1 As shown, a method for detecting facial aging includes the following steps:

[0037] Step 101: Obtain a first three-dimensional mesh model and a second three-dimensional mesh model of the target face; the first three-dimensional mesh model contains multiple first vertices, the second three-dimensional mesh model contains multiple second vertices, and the first three-dimensional mesh model and the second three-dimensional mesh model also contain multiple set facial feature points.

[0038] The target face refers to the face to be detected, that is, the specific face that needs to be detected for aging.

[0039] In some embodiments, surface geometric data of the target face are acquired using an image acquisition device, such as a depth camera, a 3D laser scanner, or a 3D structured light scanner, to obtain a 3D point cloud of the face that constitutes the target face.

[0040] The first 3D mesh model is a 3D geometric representation of the target face in a specified detection state. It contains multiple defined facial feature points, multiple first vertices, and multiple first polygonal patches. The specified detection state refers to a specific time period, such as the current state or other time period states (e.g., 40 years old).

[0041] The second three-dimensional mesh model is a three-dimensional geometric representation of the target face in a youthful state (aging reference state), which includes multiple defined facial feature points, multiple second vertices, and multiple second polygonal patches.

[0042] In this application, "youthful state" refers to a facial state at an earlier time relative to the first 3D mesh model, and is youthful relative to a specified detection state of the first 3D mesh model to be detected.

[0043] Facial feature points are fixed markers on the target face that have structural stability, such as bony landmarks (e.g., zygomatic tuberosities, mandibular angles) and muscle attachment points (e.g., corners of the eyes, corners of the mouth). These points maintain relative spatial stability during facial expression changes and slight displacements, providing a geometric benchmark for 3D modeling and aging detection, and helping to achieve spatial registration and deformation.

[0044] The first polygonal patch is a surface unit in the first 3D mesh model, formed by connecting multiple first vertices. Multiple first polygonal patches form a continuous surface mesh of the target face in a specified detection state.

[0045] The second polygonal patch is a surface unit in the second 3D mesh model, formed by connecting multiple second vertices. Multiple second polygonal patches form a continuous surface mesh of the target face in a youthful state.

[0046] In some embodiments, the first polygonal patch and the second polygonal patch are typically triangular patches or quadrilateral patches. Because triangles have topological stability and can avoid non-planar distortion, the first polygonal patch is typically defined as a first triangular patch, and the second polygonal patch is defined as a second triangular patch.

[0047] Multiple vertices of a patch contain curvature information. Obtaining a first 3D mesh model containing the first vertex and a second 3D mesh model containing the second vertex facilitates the quantization of concave and drooping features through changes in vertex curvature.

[0048] In some embodiments, obtaining the first three-dimensional mesh model and the second three-dimensional mesh model of the target face includes: performing triangulation processing on the current three-dimensional point cloud of the target face to obtain the first three-dimensional mesh model; the plurality of the set facial feature points of the first three-dimensional mesh model constitute a first feature point geometric configuration, and the plurality of the first vertices of the first three-dimensional mesh model constitute a first topological connection relationship; obtaining a young three-dimensional mesh model having the first feature point geometric configuration and the first topological connection relationship as the second three-dimensional mesh model.

[0049] The surface geometry data of the target face is acquired using an image acquisition device to obtain the current three-dimensional point cloud.

[0050] Triangulation methods such as Delaunay triangulation or Poisson surface reconstruction are used to triangulate the current 3D point cloud. To achieve spatial alignment of the model, predefined facial feature points are identified from the current 3D point cloud using deep learning feature extraction methods or local geometric curvature analysis methods. These feature points are then marked in the first 3D mesh model. Alternatively, the first 3D mesh model can be directly used as the identification object to identify and mark the predefined facial feature points, facilitating their clear identification. Through these processes, a first 3D mesh model containing multiple first vertices and multiple predefined facial feature points is finally obtained.

[0051] The first three-dimensional mesh model contains multiple set facial feature points that constitute the first feature point geometry. The set facial feature points are fixed marker points with structural stability on the target face. Correspondingly, the first feature point geometry constituted by them also has a relatively stable structure and can be used as a rigid reference system for spatial alignment.

[0052] The first vertices of the first 3D mesh model constitute a first topological connection relationship, based on which vertex correspondence can be realized. The topological connection relationship only includes the spatial distribution of vertices and the number of connections, but does not include the orientation of the connections.

[0053] In some embodiments, the geometric configuration of the first feature points and the first topological connection relationship are used as matching benchmarks to match a young 3D mesh model with the first feature point geometric configuration and the first topological connection relationship. This young 3D mesh model is then identified as the second 3D mesh model, i.e., a reference model that better matches the growth pattern of the target face. Based on the vertices in the two models, a longitudinal aging comparison of the individual at different time periods is achieved, resulting in more objective aging detection and more reliable detection results.

[0054] In some embodiments, obtaining a young 3D mesh model having the first feature point geometric configuration and the first topological connection relationship as the second 3D mesh model includes: using the first feature point geometric configuration and the first topological connection relationship as young state processing constraints to perform young state processing on the first 3D mesh model to obtain the second 3D mesh model; or, adjusting the second feature point geometric configuration of the set young 3D mesh model to the first feature point geometric configuration, and reconstructing the mesh structure of the set young 3D mesh model based on the first topological connection relationship to obtain the second 3D mesh model; or, obtaining historical 3D mesh models of the target face in a set age range from a database, and determining the historical 3D mesh model having the first feature point geometric configuration and the first topological connection relationship as the second 3D mesh model.

[0055] In some embodiments, the second three-dimensional mesh model is obtained through real-time generation or database retrieval, which improves the flexibility of obtaining the second three-dimensional mesh model.

[0056] In some embodiments, vertex smoothing is performed on the first 3D mesh model to make its surface smoother and more compact, tending towards a younger state. Vertex smoothing can include Laplacian smoothing, bilateral smoothing, etc. During vertex smoothing, the geometric configuration of the first feature point and the first topological connectivity relationship are used as constraints for the younger state processing, ensuring that the first 3D mesh model maintains these geometric configurations during the vertex smoothing process, resulting in a second 3D mesh model possessing the geometric configuration and topological connectivity relationship of the first feature point.

[0057] In some embodiments, a trained neural network model (such as an autoencoder or a generative adversarial network) can be used to predict and generate a young-state 3D mesh model corresponding to the first 3D mesh model. During the prediction and generation process, the geometric configuration of the first feature points and the first topological connectivity relationship also need to be used as constraints for young-state processing, so that the first 3D mesh model maintains the geometric configuration of the first feature points and the first topological connectivity relationship during neural network processing, resulting in a second 3D mesh model possessing the geometric configuration of the first feature points and the first topological connectivity relationship.

[0058] In some embodiments, the young state 3D mesh model is set as a pre-stored standardized young state model, or the young state 3D mesh model is set as a standardized young state model generated in real time using the Statistical Shape Model (SSM) or Principal Component Analysis (PCA) method.

[0059] In some embodiments, multiple defined facial feature points of the youthful 3D mesh model are adjusted, and the geometric configuration of the second feature points, composed of these multiple defined facial feature points, is adjusted to the geometric configuration of the first feature points. This adjustment can be achieved through singular value decomposition or interpolation methods to ensure that the geometric configuration of the feature points of the youthful model adapts to the stable architecture of the target face. Simultaneously, the second topological connection relationship of the youthful 3D mesh model is adjusted based on the first topological connection relationship; that is, the mesh connection structure is reconstructed, or the faces are re-divided, ensuring that the number of vertices and connection relationships are consistent with the first 3D mesh model. This facilitates vertex mapping and ensures that each first vertex has a corresponding reference vertex. Here, the first feature point geometric configuration and the first topological connection relationship are constraints. Through feature point adjustment and mesh reconstruction, a second 3D mesh model with the first feature point geometric configuration and the first topological connection relationship is obtained.

[0060] In some embodiments, based on the individual information corresponding to the target face, a historical 3D mesh model of the target face within a set age range is obtained from a database. The age corresponding to the set age range is less than the age corresponding to the first 3D mesh model. The geometric configuration of the third feature point of the obtained historical 3D mesh model is compared with the geometric configuration of the first feature point, and the corresponding third topological connection relationship is compared with the first topological connection relationship. If they are consistent, the historical 3D mesh model is determined as the second 3D mesh model. If the geometric configuration of the feature points is inconsistent and / or the topological connection relationship is inconsistent, the geometric configuration of the third feature point and / or the third topological connection relationship of the historical 3D mesh model is adjusted to obtain a historical 3D mesh model with the geometric configuration of the first feature point and the first topological connection relationship, and this is determined as the second 3D mesh model. In this way, even if the target face has undergone cosmetic surgery, a suitable reference model can be matched according to the actual situation.

[0061] In some embodiments, the young three-dimensional mesh model can also be a model selected by the operator. In order to achieve spatial registration and deformation, it will be processed to give it a first feature point geometric configuration and a first topological connection relationship, thereby obtaining a second three-dimensional mesh model.

[0062] In some embodiments, aging detection can be performed on a specified first 3D mesh model that is not in its current state. That is, a pre-stored first 3D mesh model is obtained, and based on the first feature point geometry and first topological connectivity of the first 3D mesh model, a second 3D mesh model is matched as a detection reference benchmark to enhance cross-time comparability.

[0063] Step 102: Based on the multiple set facial feature points, align the first three-dimensional mesh model and the second three-dimensional mesh model to the same spatial coordinate system, and establish a one-to-one vertex mapping relationship between multiple first vertices and multiple second vertices accordingly.

[0064] After alignment, the first and second 3D mesh models share the same spatial coordinate system, which can accurately quantify the changes in vertex positions, thereby improving quantization efficiency and accuracy.

[0065] In some embodiments, based on multiple set facial feature points, the first three-dimensional mesh model is aligned to the spatial coordinate system of the second three-dimensional mesh model; or, based on multiple set facial feature points, the second three-dimensional mesh model is aligned to the spatial coordinate system of the first three-dimensional mesh model; or, based on multiple set facial feature points, the first three-dimensional mesh model and the second three-dimensional mesh model are aligned to a set spatial coordinate system such as the world coordinate system.

[0066] In some embodiments, the rigid transformation matrix between the first 3D mesh model and the second 3D mesh model is solved, so that the defined facial feature points of the first 3D mesh model and the second 3D mesh model coincide in the same coordinate system or are spaced apart by a defined distance. Spatial alignment of the first 3D mesh model and the second 3D mesh model is achieved based on this rigid transformation matrix.

[0067] In some embodiments, the topological connections of the first 3D mesh model and the second 3D mesh model are consistent. For each first vertex in the first 3D mesh model, a second vertex with the same topological connection and the closest distance to that first vertex is searched in the second 3D mesh model. A correspondence between the two vertices is established, ultimately resulting in a one-to-one vertex mapping relationship between multiple first vertices and multiple second vertices. Then, using the second vertex as a reference point, the positional change of the corresponding first vertex is determined.

[0068] Step 103: Calculate the projection difference of each first vertex relative to its corresponding second vertex.

[0069] In the same spatial coordinate system, calculate the projection difference of the first vertex relative to its corresponding second vertex, that is, the change in position of the first vertex, so as to assess the aging of the first vertex based on the change in position.

[0070] In some embodiments, calculating the projection difference between each first vertex and its corresponding second vertex includes: calculating a difference vector from the second vertex to the first vertex based on the first three-dimensional coordinates of each first vertex and the second three-dimensional coordinates of its corresponding second vertex; and performing a dot product operation on the difference vector and the unit normal vector at the second vertex to obtain the projection difference.

[0071] In some embodiments, the first three-dimensional mesh model is The second three-dimensional mesh model is .in, The first vertex set contains multiple first vertices. ; This is the set of second vertices, containing multiple second vertices. , It is a positive integer; A set of polygon patches contains multiple polygon patches. If two models contain the same set of polygon patches, it means that the topological connections between the models are consistent.

[0072] First Vertex Set The first vertex in With the second vertex set The second vertex A vertex mapping relationship exists. The first three-dimensional coordinates are , The second three-dimensional coordinates are . the following Reference , Reference Second vertex Pointing to the first vertex difference vector .

[0073] Get the second vertex Normal vector at point First vertex projection difference .

[0074] In addition to the methods described above, other mathematical methods can be used to calculate the projection difference between each first vertex and its corresponding second vertex. This application does not limit the method of calculation.

[0075] Step 104: Determine the concavity information and drooping information of the multiple first vertices based on the multiple projection differences.

[0076] After obtaining the projection difference corresponding to each first vertex, the concavity and drooping information of each first vertex are evaluated based on the projection difference to realize vertex aging determination, that is, to realize aging detection of each local area of ​​the target face.

[0077] The concavity information of the first vertex includes whether a concavity feature exists and, if so, the specific concavity level. The drooping information of the first vertex includes whether a drooping feature exists and, if so, the specific drooping level. In other words, the aging determination result of the first vertex is divided into three categories: no concavity or drooping features; concavity features exist; and drooping features exist. Each first vertex corresponds to only one of these three results. Multiple first vertices can be further distinguished based on the concavity and drooping levels.

[0078] In some embodiments, determining the concavity information and drooping information of a plurality of first vertices based on a plurality of projection differences includes: determining the maximum difference and the minimum difference among the plurality of projection differences; normalizing the plurality of projection differences based on the maximum difference and the minimum difference to obtain a target difference for each first vertex; comparing the target difference for each first vertex with a plurality of preset thresholds to obtain a comparison result; and determining the presence of concavity features and drooping features for each first vertex based on the comparison result, and determining the concavity level of the first vertex with the concavity feature and the drooping level of the first vertex with the drooping feature.

[0079] By eliminating dimensional differences through normalization and combining multi-level threshold dynamic comparison, the automated identification and quantitative evaluation of the concave and sagging features on the surface of the first three-dimensional mesh model were achieved.

[0080] In some embodiments, the normalization process is implemented based on multiple calculated projection differences. The maximum and minimum differences among the multiple projection differences are determined. The target difference for each first vertex is determined. ,in, To be the minimum difference, The maximum difference is calculated. Normalization linearly maps the original data to a specified interval, such as [0, 1] or [-1, 1], eliminating dimensional differences, unifying the data scale, and facilitating subsequent comparative analysis.

[0081] Multiple preset thresholds are used to compare with the target difference to determine the indentation and sagging information corresponding to the target difference.

[0082] In some embodiments, the plurality of set thresholds include a first set threshold, a second set threshold, a third set threshold, a fourth set threshold, a fifth set threshold, and a sixth set threshold, wherein the first set threshold is greater than the second set threshold, the second set threshold is greater than the third set threshold, the third set threshold is greater than the fourth set threshold, the fourth set threshold is less than the fifth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0083] Accordingly, determining the presence of concavity features and drooping features of each first vertex based on the comparison results, and determining the concavity level of the first vertex exhibiting the concavity feature and the drooping level of the first vertex exhibiting the drooping feature, includes:

[0084] If the target difference is greater than or equal to the first set threshold, then the first vertex is determined to have drooping characteristics and the drooping level is severe drooping.

[0085] If the target difference is less than the first set threshold and greater than or equal to the second set threshold, then it is determined that the first vertex has a drooping feature and the drooping level is moderate drooping.

[0086] If the target difference is less than the second set threshold and greater than the third set threshold, then it is determined that the first vertex has a drooping feature and the drooping level is mild drooping.

[0087] If the target difference is less than or equal to the third set threshold and greater than or equal to the fourth set threshold, then it is determined that the first vertex does not have drooping or concave features.

[0088] If the target difference is less than the fourth set threshold and greater than the fifth set threshold, then it is determined that the first vertex has a concave feature and the concave level is mild concavity.

[0089] If the target difference is less than or equal to the fifth set threshold and greater than the sixth set threshold, then the first vertex is determined to have a concave feature and the concave level is moderate concavity.

[0090] If the target difference is less than or equal to the sixth set threshold, then the first vertex is determined to have a concave feature and the concave level is severe concavity.

[0091] In some embodiments, the plurality of set thresholds include a seventh set threshold, an eighth set threshold, a ninth set threshold, and a tenth set threshold, wherein the seventh set threshold is greater than the eighth set threshold, the eighth set threshold is greater than the ninth set threshold, and the ninth set threshold is greater than the tenth set threshold.

[0092] Accordingly, determining the presence of concavity features and drooping features of each first vertex based on the comparison results, and determining the concavity level of the first vertex exhibiting the concavity feature and the drooping level of the first vertex exhibiting the drooping feature, includes:

[0093] If the target difference is greater than or equal to the seventh set threshold, then it is determined that the first vertex has a drooping feature and the drooping level is level two drooping.

[0094] If the target difference is less than the seventh set threshold and greater than the eighth set threshold, then it is determined that the first vertex has a drooping feature and the drooping level is level one drooping.

[0095] If the target difference is less than or equal to the eighth set threshold and greater than or equal to the ninth set threshold, then it is determined that the first vertex does not have drooping or concave features.

[0096] If the target difference is less than the ninth set threshold and greater than the tenth set threshold, then it is determined that the first vertex has a concave feature and the concave level is a first-level concave.

[0097] If the target difference is less than or equal to the tenth set threshold, then the first vertex is determined to have a concave feature and the concave level is a level two concave.

[0098] Among them, the degree of sag in secondary sag is greater than that in primary sag, and the degree of depression in secondary depression is greater than that in primary depression.

[0099] In practical applications, different numbers and sizes of threshold values ​​can be set as needed to achieve aging determination under different aging quantification standards.

[0100] Step 105: Based on the concavity and drooping information of multiple first vertices, generate and output an aging detection report of the target face.

[0101] By integrating scattered vertex-level depression and sag information into a structured report, the system transforms local features into a global aging assessment. The generation and output of this aging detection report can assist in medical aesthetics and health management.

[0102] In some embodiments, the feature existence and level information of multiple first vertices can be directly output as the report information of the aging detection report, that is, the aging detection report includes the concavity information and drooping information of multiple first vertices.

[0103] In some embodiments, to facilitate users' more intuitive understanding of the aging of the target face, more detailed data visualization processing can be performed based on the concavity and drooping information of multiple first vertices.

[0104] In some embodiments, the aging detection report includes image information and an aging score. Generating and outputting the aging detection report of the target face based on the concavity and drooping information of multiple first vertices includes: using different colors to mark first vertices with different concavity levels, first vertices with different drooping levels, and first vertices without concavity or drooping features in the first three-dimensional mesh model, generating an aging distribution three-dimensional chromatographic model of the target face; determining the area proportion of each color-marked region in the aging distribution three-dimensional chromatographic model; calculating the comprehensive aging score of the target face based on the area proportion of each color-marked region and the scoring weight coefficients corresponding to each color; and outputting the aging detection report including the aging distribution three-dimensional chromatographic model and the comprehensive aging score.

[0105] The system simultaneously outputs a three-dimensional chromatographic model of aging distribution and a comprehensive aging score, forming a closed-loop report of images and data. This helps in understanding the test results, facilitates comparisons of aging between different individuals and at different time points within the same individual, and the rich report information also avoids subjective bias from human interpretation.

[0106] In some embodiments, given the concavity and drooping information of each first vertex, when generating a three-dimensional colorimetric model of the aging distribution of the target face, the rendering color is determined based on the concavity level and the drooping level of each first vertex; and the rendering area is determined based on the target difference of each first vertex.

[0107] In some embodiments, the rendering color of the first vertex with a concave feature is a warm color, the rendering color of the first vertex with a drooping feature is a cool color, and the rendering color of the first vertex without a concave or drooping feature is a neutral color.

[0108] In some embodiments, warm hues of different saturations indicate first vertices of different depression levels, and cool hues of different saturations indicate first vertices of different sagging levels.

[0109] In some embodiments, a severely drooping first vertex is marked in red, a moderately drooping first vertex is marked in orange, a slightly drooping first vertex is marked in yellow, a first vertex without any concave or drooping features is marked in green, a slightly concave first vertex is marked in cyan, a moderately concave first vertex is marked in blue, and a severely concave first vertex is marked in purple.

[0110] In some embodiments, the rendering region area is determined based on the target difference of the first vertex. The rendering region area is applied to the color rendering of the first vertex, which has concave or drooping features. In a specific implementation, vertex region rendering is performed with the first vertex as the rendering center point, according to the determined rendering color and the rendering region area corresponding to the target difference.

[0111] In some embodiments, for the first vertex with a droop level, the target difference is represented in the form of "positive sign + value", such as "+0.82". The larger the target difference, the larger the rendering area; that is, the larger the value after the positive sign, the larger the rendering area. For the first vertex with a concave level, the target difference is represented in the form of "negative sign + value", such as "-0.57". The smaller the target difference, the larger the rendering area; that is, the larger the value after the negative sign, the larger the rendering area.

[0112] In some embodiments, the first vertex in the first three-dimensional mesh model that does not have concave or drooping features, as well as other unrendered areas, are rendered using the same color (i.e., a color distinct from the rendering color of the first vertex with concave or drooping features), such as green.

[0113] By using color mapping, the abstract aging features of three-dimensional mesh vertices are transformed into an intuitive three-dimensional color spectrum representation, enabling regional visualization and localization of aging features.

[0114] like Figure 2 As shown, Figure 2 This is a schematic diagram of a three-dimensional chromatographic model of aging distribution provided in an embodiment of this application. Figure 2 The three-dimensional mesh model shown uses multiple colors to indicate the aging distribution. Figure 2 The 3D face models involved are non-realistic face models generated based on technologies such as artificial intelligence or deep learning.

[0115] In practical applications, users can adjust the viewing angle of the 3D chromatographic model of aging distribution through view controls to view the aging status of the target face from different perspectives. Figure 2 The three-dimensional chromatographic model of aging distribution is shown from different perspectives, where (a), (b), and (c) correspond to the left-side view, front view, and right-side view, respectively. (a), (b), and (c) are the left-side view, front view, and right-side view of the three-dimensional chromatographic model of aging distribution, respectively.

[0116] exist Figure 2In this study, warm colors (red and yellow), neutral colors (green), and cool colors (dark blue and light blue) with varying saturations are used to indicate facial areas at different stages of aging. Green areas represent facial areas without depressions or sagging; red areas represent facial areas with secondary depressions; yellow areas represent facial areas with primary depressions; light blue areas represent facial areas with primary sagging; and dark blue areas represent facial areas with secondary sagging.

[0117] based on Figure 2 It can clearly and intuitively understand the aging state of the target face, thereby achieving targeted rejuvenation treatment.

[0118] In some embodiments, the image visualization method is not limited to color mapping, but also supports visualization mapping in the form of transparency, texture, etc.

[0119] In some embodiments, the area proportion of each color-coded region in the 3D aging distribution chromatographic model is statistically analyzed. Simultaneously, the scoring weight coefficients corresponding to the marked colors of each region are combined, and a weighted summation is performed to calculate the overall aging score of the target face. This overall aging score represents the global aging level of the target face. Subsequently, an aging detection report containing the 3D aging distribution chromatographic model and the overall aging score is output. The 3D aging distribution chromatographic model and the overall aging score objectively reflect the degree of aging of the target face, which helps in determining effective cosmetic medical solutions. For example, based on the 3D aging distribution chromatographic model, targeted cosmetic medical interventions can be performed on specific regions of the target face.

[0120] In some embodiments, the aging detection report also includes information such as the proportion of different levels of features and suggestions for medical aesthetic adjustments, transforming discrete geometric features into actionable aging assessment conclusions, which has strong practicality in scenarios such as medical aesthetics and health management.

[0121] In this embodiment, based on the set facial feature points included in both the first and second 3D mesh models, the first and second 3D mesh models are aligned to the same spatial coordinate system. A one-to-one vertex mapping relationship is established between multiple first vertices of the first 3D mesh model and multiple second vertices of the second 3D mesh model within this spatial coordinate system. This allows for the calculation of the projection difference between each first vertex and its corresponding second vertex, quantifying the concavity and drooping information of each first vertex. This generates and outputs an aging detection report for the target face, providing a face aging detection method based on concavity and drooping features. This fills the gap in traditional methods for 3D structural aging detection and solves the problem in existing technologies that do not combine concavity and drooping features to achieve face aging detection.

[0122] See Figure 3 , Figure 3 This is a structural diagram of a face aging detection system provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0123] The facial aging detection system 300 includes: an acquisition module 301, an alignment and mapping module 302, a calculation module 303, a determination module 304, and an output generation module 305.

[0124] The acquisition module 301 is used to acquire a first three-dimensional mesh model and a second three-dimensional mesh model of the target face; the first three-dimensional mesh model contains a plurality of first vertices, the second three-dimensional mesh model contains a plurality of second vertices, and the first three-dimensional mesh model and the second three-dimensional mesh model also contain a plurality of set facial feature points.

[0125] The alignment mapping module 302 is used to align the first three-dimensional mesh model and the second three-dimensional mesh model to the same spatial coordinate system based on multiple set facial feature points, and to establish a one-to-one vertex mapping relationship between multiple first vertices and multiple second vertices accordingly.

[0126] The calculation module 303 is used to calculate the projection difference between each of the first vertices and its corresponding second vertex.

[0127] The determining module 304 is used to determine the concavity information and drooping information of the plurality of first vertices based on the plurality of projection differences.

[0128] The output generation module 305 is used to generate and output an aging detection report of the target face based on the concavity information and drooping information of multiple first vertices.

[0129] In some embodiments, the acquisition module is specifically used for:

[0130] The current three-dimensional point cloud of the target face is triangulated to obtain a first three-dimensional mesh model; the multiple set facial feature points of the first three-dimensional mesh model constitute a first feature point geometric configuration, and the multiple first vertices of the first three-dimensional mesh model constitute a first topological connection relationship.

[0131] A young state 3D mesh model with the first feature point geometric configuration and the first topological connection relationship is obtained as the second 3D mesh model.

[0132] In some embodiments, the acquisition module is further configured to:

[0133] Using the geometric configuration of the first feature points and the first topological connectivity as constraints for young state processing, the first 3D mesh model is subjected to young state processing to obtain the second 3D mesh model; or...

[0134] The geometric configuration of the second feature point of the young state 3D mesh model is adjusted to the geometric configuration of the first feature point, and the mesh structure of the young state 3D mesh model is reconstructed based on the first topological connectivity relationship to obtain the second 3D mesh model; or,

[0135] The historical 3D mesh model of the target face in a set age range is obtained from the database, and the historical 3D mesh model with the first feature point geometric configuration and the first topological connection relationship is determined as the second 3D mesh model.

[0136] In some embodiments, the computing module is specifically used for:

[0137] Based on the first three-dimensional coordinates of each first vertex and the corresponding second three-dimensional coordinates of the second vertex, calculate the difference vector from the second vertex to the first vertex;

[0138] Perform a dot product operation on the difference vector and the unit normal vector at the second vertex to obtain the projection difference.

[0139] In some embodiments, the determining module is specifically used for:

[0140] Determine the maximum and minimum difference among the plurality of projection differences;

[0141] Based on the maximum difference and the minimum difference, the multiple projection differences are normalized to obtain the target difference of each first vertex;

[0142] The target difference of each first vertex is compared with multiple set thresholds to obtain a comparison result;

[0143] Based on the comparison results, the presence of concave features and drooping features of each first vertex are determined, and the concave level of the first vertex with the concave features and the drooping level of the first vertex with the drooping features are determined.

[0144] In some embodiments, the output generation module is specifically used for:

[0145] The first vertices of different concavity levels, different drooping levels, and the first vertices without concavity and drooping features in the first three-dimensional mesh model are marked with different colors to generate a three-dimensional color chromatogram model of the aging distribution of the target face.

[0146] Determine the area proportion of each color-coded region in the three-dimensional chromatographic model of aging distribution;

[0147] Based on the area proportion of each color-marked region and the scoring weight coefficient corresponding to each color, the comprehensive aging score of the target face is calculated.

[0148] The output includes the three-dimensional chromatographic model of the aging distribution and the comprehensive aging score, forming the aging detection report.

[0149] In some embodiments, the output generation module is further configured to:

[0150] The rendering color is determined based on the concavity level and the droop level of each of the first vertices;

[0151] The area of ​​the rendering region is determined based on the target difference for each of the first vertices.

[0152] The facial aging detection system provided in this application embodiment can implement all the processes of the above-described facial aging detection method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0153] Figure 4 This is a structural diagram of an electronic device provided in an embodiment of this application. As shown in the figure, the electronic device 4 of this embodiment includes: at least one processor 40 ( Figure 4 (Only one is shown in the diagram), memory 41, and computer program 42 stored in said memory 41 and executable on said at least one processor 40, which, when executed, implements the steps in any of the above method embodiments.

[0154] The electronic device 4 can be a desktop computer, laptop, handheld computer, or cloud server, etc. The electronic device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0155] The processor 40 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0156] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard disk or memory. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 41 can include both internal and external storage units of the electronic device 4. The memory 41 is used to store the computer program and other programs and data required by the electronic device. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0157] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0158] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0159] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0160] In the embodiments provided in this application, it should be understood that the disclosed systems / electronic devices and methods can be implemented in other ways. For example, the system / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0161] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0162] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0163] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0164] The processes in the above-described embodiments can be implemented by a computer program product. When the computer program product is run on an electronic device, the electronic device executes the steps in the above-described method embodiments.

[0165] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting facial aging, characterized in that, include: Obtain a first 3D mesh model and a second 3D mesh model of the target face; the first 3D mesh model contains multiple first vertices, the second 3D mesh model contains multiple second vertices, and the first 3D mesh model and the second 3D mesh model also contain multiple defined facial feature points; Based on multiple defined facial feature points, the first three-dimensional mesh model and the second three-dimensional mesh model are aligned to the same spatial coordinate system, and a one-to-one vertex mapping relationship is established between multiple first vertices and multiple second vertices accordingly. Calculate the projection difference of each first vertex relative to its corresponding second vertex; Based on multiple projection differences, the concavity information and drooping information of multiple first vertices are determined, specifically including: determining the maximum and minimum differences among the multiple projection differences; normalizing the multiple projection differences based on the maximum and minimum differences to obtain target differences for each first vertex; comparing the target difference of each first vertex with multiple preset thresholds for determining concavity features and multiple preset thresholds for determining drooping features to obtain comparison results; and determining the presence of concavity features and drooping features for each first vertex based on the comparison results, and determining the concavity level of the first vertex with the concavity features and the drooping level of the first vertex with the drooping features. Based on the concavity and drooping information of multiple first vertices, an aging detection report of the target face is generated and output.

2. The method according to claim 1, characterized in that, The acquisition of the first and second 3D mesh models of the target face includes: The current three-dimensional point cloud of the target face is triangulated to obtain a first three-dimensional mesh model; the multiple set facial feature points of the first three-dimensional mesh model constitute a first feature point geometric configuration, and the multiple first vertices of the first three-dimensional mesh model constitute a first topological connection relationship. A young state 3D mesh model with the first feature point geometric configuration and the first topological connection relationship is obtained as the second 3D mesh model.

3. The method according to claim 2, characterized in that, The step of obtaining a young-state 3D mesh model having the first feature point geometric configuration and the first topological connection relationship as the second 3D mesh model includes: Using the geometric configuration of the first feature points and the first topological connectivity as constraints for young state processing, the first 3D mesh model is subjected to young state processing to obtain the second 3D mesh model; or... The geometric configuration of the second feature point of the young state 3D mesh model is adjusted to the geometric configuration of the first feature point, and the mesh structure of the young state 3D mesh model is reconstructed based on the first topological connectivity relationship to obtain the second 3D mesh model; or, The historical 3D mesh model of the target face in a set age range is obtained from the database, and the historical 3D mesh model with the first feature point geometric configuration and the first topological connection relationship is determined as the second 3D mesh model.

4. The method according to claim 1, characterized in that, The calculation of the projection difference between each of the first vertex and its corresponding second vertex includes: Based on the first three-dimensional coordinates of each first vertex and the corresponding second three-dimensional coordinates of the second vertex, calculate the difference vector from the second vertex to the first vertex; Perform a dot product operation on the difference vector and the unit normal vector at the second vertex to obtain the projection difference.

5. The method according to claim 1, characterized in that, The process of generating and outputting an aging detection report for the target face based on the concavity and drooping information of multiple first vertices includes: The first vertices of different concavity levels, different drooping levels, and the first vertices without concavity and drooping features in the first three-dimensional mesh model are marked with different colors to generate a three-dimensional color chromatogram model of the aging distribution of the target face. Determine the area proportion of each color-coded region in the three-dimensional chromatographic model of aging distribution; Based on the area proportion of each color-marked region and the scoring weight coefficient corresponding to each color, the comprehensive aging score of the target face is calculated. The output includes the three-dimensional chromatographic model of the aging distribution and the comprehensive aging score, forming the aging detection report.

6. The method according to claim 5, characterized in that, The method further includes: The rendering color is determined based on the concavity level and the droop level of each of the first vertices; The area of ​​the rendering region is determined based on the target difference for each of the first vertices.

7. A facial aging detection system, characterized in that, include: The acquisition module is used to acquire a first three-dimensional mesh model and a second three-dimensional mesh model of the target face; the first three-dimensional mesh model contains multiple first vertices, the second three-dimensional mesh model contains multiple second vertices, and the first three-dimensional mesh model and the second three-dimensional mesh model also contain multiple set facial feature points; The alignment mapping module is used to align the first three-dimensional mesh model and the second three-dimensional mesh model to the same spatial coordinate system based on multiple set facial feature points, and to establish a one-to-one vertex mapping relationship between multiple first vertices and multiple second vertices accordingly. The calculation module is used to calculate the projection difference of each first vertex relative to its corresponding second vertex; The determining module is configured to determine the concavity information and drooping information of a plurality of first vertices based on a plurality of projection differences. Specifically, this includes: determining the maximum and minimum difference among the plurality of projection differences; normalizing the plurality of projection differences based on the maximum and minimum difference to obtain a target difference for each first vertex; comparing the target difference for each first vertex with a plurality of preset thresholds for determining concavity features and a plurality of preset thresholds for determining drooping features to obtain a comparison result; and determining the presence of concavity features and drooping features for each first vertex based on the comparison result, and determining the concavity level of the first vertex exhibiting the concavity feature and the drooping level of the first vertex exhibiting the drooping feature. The output generation module is used to generate and output an aging detection report of the target face based on the concavity and drooping information of multiple first vertices.

8. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device performs the method as described in any one of claims 1 to 6.

9. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method as described in any one of claims 1 to 6 to be performed.

Images