Data verification method, computer device, and readable storage medium

Abstract

Embodiments of the present application disclose a data verification method, a computer device, and a readable storage medium, relating to the image recognition technology in artificial intelligence. Among them, the method includes: in response to a transformation trigger instruction for a transformation function, identifying an initial marking point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction; performing transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data; the transformation trigger instruction includes an expected transformation value for the transformation processing; identifying an updated marking point corresponding to the initial marking point in the updated multimedia data; obtaining a marking difference value between the initial marking point and the updated marking point; and determining a detection result of the transformation function according to the marking difference value and the expected transformation value. By using the embodiments of the present application, it is possible to detect the transformation function and improve the accuracy of data verification.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular, to a data verification method, a computer device, and a readable storage medium. Background Art

[0002] In the era of using multimedia data for social interaction, users hope to show their beautiful appearance and form on social platforms. Therefore, users need to use application programs that can transform the multimedia data obtained by taking pictures of themselves to beautify the multimedia data, so as to improve user satisfaction with the beautified multimedia data.

[0003] In the prior art, due to the large number and variety of such application programs and uneven transformation functions, how to detect the transformation functions of these application programs and determine whether the transformation functions of the application programs are normal, so as to improve the accuracy of data verification is an urgent problem to be solved. Currently, generally, it is judged whether the transformation function is perfect based on the code, or whether the business logic of the transformation function is correct is judged from the user interface (UI) level. There is no clear detection method for the actual execution result of the transformation function, so when the effect of the transformation function is abnormal, it may not be possible to check, resulting in a low accuracy of data detection for the transformation function. Summary of the Invention

[0004] Embodiments of this application provide a data verification method, a computer device, and a readable storage medium, which can detect the transformation function of an application program and determine whether the transformation function is normal, thereby improving the accuracy of data verification.

[0005] One aspect of the embodiments of this application provides a data verification method, including:

[0006] In response to a transformation trigger instruction for a transformation function, identifying an initial marking point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction;

[0007] Performing a transformation process on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data; the transformation trigger instruction includes a transformation value for the transformation process;

[0008] Identifying an updated marking point corresponding to the initial marking point in the updated multimedia data;

[0009] Obtaining a marking difference value between the initial marking point and the updated marking point;

[0010] Determining a detection result of the transformation function according to the marking difference value and the expected transformation value.

[0011] One aspect of the embodiments of the present application provides a data verification device, including:

[0012] A first recognition module, configured to recognize an initial marking point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction in response to the transformation trigger instruction for the transformation function;

[0013] An object transformation module, configured to perform transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data; the transformation trigger instruction includes an expected transformation value for the transformation processing;

[0014] A second recognition module, configured to recognize an updated marking point corresponding to the initial marking point in the updated multimedia data;

[0015] A difference acquisition module, configured to acquire a marking difference value between the initial marking point and the updated marking point;

[0016] A result acquisition module, configured to determine a detection result of the transformation function according to the marking difference value and the expected transformation value.

[0017] Optionally, the first recognition module includes:

[0018] A first object determination unit, configured to recognize the multimedia data targeted by the transformation trigger instruction based on a target detection model to determine the target object in the multimedia data;

[0019] A feature determination unit, configured to acquire the object feature of the target object in the multimedia data, and determine the initial marking point corresponding to the target object from the marking points corresponding to the object feature.

[0020] Optionally, the first recognition module includes:

[0021] A parameter acquisition unit, configured to acquire target marking parameters associated with the transformation function corresponding to the transformation trigger instruction, and acquire a target marking parameter range corresponding to the target marking parameters;

[0022] An initial data determination unit, configured to determine, in the multimedia data targeted by the transformation trigger instruction, pixel points whose pixel parameters belong to the target marking parameter range as the initial marking points corresponding to the target object in the multimedia data; initial marking points with the same target marking parameters are initial marking point pairs.

[0023] Optionally, the device further includes:

[0024] A mapping module, configured to acquire pixel values of at least two pixel points in the multimedia data, and map the at least two pixel points to a color map based on the pixel values of the at least two pixel points;

[0025] A clustering module, configured to perform clustering processing on the at least two pixel points based on the positions of the at least two pixel points in the color map, so as to obtain a multimedia color cluster;

[0026] A range determination module, configured to determine a marking parameter range according to the color map values in the color map whose color distances from the multimedia color cluster are greater than a color difference threshold, where the marking parameter range includes a target marking parameter range.

[0027] Optionally, the number of the initial marking points is at least two, and the number of the updated marking points is at least two; the result acquisition module includes:

[0028] An initial data grouping unit, configured to obtain the initial parameters respectively corresponding to each of the at least two initial marking points, obtain the parameter types of the initial parameters respectively corresponding to each of the at least two initial marking points, and perform grouping processing on the at least two initial marking points based on the parameter types of the initial parameters, so as to obtain m pairs of initial marking points; the m pairs of initial marking points include the initial marking point pair i, the initial marking point pair i includes a first initial marking point and a second initial marking point, m is a positive integer, and i is a positive integer;

[0029] An updated data grouping unit, configured to obtain the updated parameters respectively corresponding to each of the at least two updated marking points, obtain the parameter types of the updated parameters respectively corresponding to each of the at least two updated marking points, and perform grouping processing on the at least two updated marking points based on the parameter types of the updated parameters, so as to obtain m pairs of updated marking points; the m pairs of updated marking points include the updated marking point pair i, the updated marking point pair i includes a first updated marking point and a second updated marking point;

[0030] An initial distance determination unit, configured to obtain the first initial pixel position of the first initial marking point in the updated multimedia data and the second initial pixel position of the second initial marking point in the updated multimedia data for each initial marking point pair, and determine the initial distance of the initial marking point pair i according to the first initial pixel position and the second initial pixel position;

[0031] An updated distance determination unit, configured to obtain the first updated pixel position of the first updated marking point in the multimedia data and the second updated pixel position of the second updated marking point in the multimedia data for each updated marking point pair, and determine the updated distance of the updated marking point pair i according to the first updated pixel position and the second updated pixel position;

[0032] A first difference determination unit, configured to determine the distance difference between the initial distance of the initial marking point pair i and the updated distance of the updated marking point pair i as the i-th marking difference value.

[0033] Optionally, the number of the initial marker points is at least two, and the number of the updated marker points is at least two; the result acquisition module includes:

[0034] An initial data matching unit, configured to obtain the target parameter type corresponding to the transformation function, obtain the parameter type of each of the at least two initial marker points, and determine the initial marker points that match the target parameter type among the at least two initial marker points as target initial marker points; the target initial marker points include a target initial marker point j and a target initial marker point k, and the parameter types of the target initial marker point j and the target initial marker point k are the same;

[0035] An initial position determination unit, configured to obtain a third initial pixel position of the target initial marker point j in the multimedia data and a fourth initial pixel position of the target initial marker point k in the multimedia data, and determine a target initial distance between the target initial marker point j and the target initial marker point k according to the third initial pixel position and the fourth initial pixel position;

[0036] An updated data matching unit, configured to obtain the parameter type of each of the at least two updated marker points, and determine the updated marker points that match the target parameter type among the at least two updated marker points as target updated marker points; the target updated marker points include a target updated marker point j and a target updated marker point k, and the parameter types of the target updated marker point j and the target updated marker point k are the same;

[0037] An updated position determination unit, configured to obtain a third updated pixel position of the target updated marker point j in the updated multimedia data and a fourth updated pixel position of the target updated marker point k in the updated multimedia data, and determine a target updated distance between the target updated marker point j and the target updated marker point k according to the third updated pixel position and the fourth updated pixel position;

[0038] A second difference determination unit, configured to determine a distance difference between the target initial distance and the target updated distance as the marker difference value.

[0039] Optionally, the result acquisition module includes:

[0040] A difference data acquisition unit, configured to obtain difference data between the marker difference value and the expected transformation value;

[0041] A first result determination unit, configured to determine that the detection result of the transformation function is a valid transformation function result if the difference data belongs to a valid error range;

[0042] A second result determination unit, configured to determine that the detection result of the transformation function is an invalid transformation function result if the difference data does not belong to the valid error range.

[0043] Optionally, the apparatus further includes:

[0044] A request generation module, configured to generate a function repair request for the transformation function if the detection result of the transformation function is an invalid result of the transformation function; the function repair request includes the difference data;

[0045] A request sending module, configured to send the function repair request to a user terminal, so that the user terminal repairs the transformation function based on the function repair request.

[0046] On the one hand, the present application provides a computer device, including: a processor, a memory, and a network interface;

[0047] The above-mentioned processor is connected to the memory and the network interface. Among them, the network interface is used to provide a data communication function, the above-mentioned memory is used to store a computer program, and the above-mentioned processor is used to call the above-mentioned computer program to execute the method in the above-mentioned one aspect in the embodiments of the present application.

[0048] On the one hand, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored, and the computer program is adapted to be loaded and executed by a processor to execute the method in the above-mentioned first aspect.

[0049] On the one hand, an embodiment of the present application provides a computer program product or a computer program, the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various alternative manners in one aspect of the embodiments of the present application.

[0050] In the embodiments of the present application, in response to a transformation trigger instruction for a transformation function, an initial marking point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction is identified; the target object in the multimedia data is subjected to transformation processing based on the transformation function to obtain updated multimedia data; wherein, the transformation trigger instruction includes an expected transformation value for the transformation processing; an updated marking point corresponding to the initial marking point in the updated multimedia data is identified; a marking difference value between the initial marking point and the updated marking point is obtained, and based on the marking difference value and the expected transformation value, a detection result of the transformation function is determined. Since the expected transformation value included in the transformation trigger instruction can represent the theoretical transformation data volume that the transformation function can transform the multimedia data, and the marking difference value can represent the actual changed data volume of the multimedia data after the transformation processing, by comparing the theoretical transformation data volume and the actual changed data volume of the multimedia data, the detection result of the transformation function can be determined, that is, it can be determined whether the transformation function is effective or invalid; since the marking points in the multimedia data before and after the transformation are identified, and the marking difference value between the initial marking point and the updated marking point, as well as the difference between the marking difference value and the expected transformation value are compared, the accuracy of data verification can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0052] Figure 1 is a schematic diagram of the architecture of a data verification system provided by an embodiment of the present application;

[0053] Figure 2 is a schematic diagram of an application scenario of a data verification method provided by an embodiment of the present application;

[0054] Figure 3 is a schematic flowchart of a data verification method provided by an embodiment of the present application;

[0055] Figure 4a is a schematic diagram of determining a marking parameter provided by an embodiment of the present application;

[0056] Figure 4b is a schematic diagram of a scenario for determining a marking parameter range provided by an embodiment of the present application;

[0057] Figure 5 is a schematic diagram of transforming a target object provided by an embodiment of the present application;

[0058] Figure 6It is a schematic flowchart of a data verification method provided by an embodiment of the present application;

[0059] Figure 7 It is a schematic diagram of the composition structure of a data verification device provided by an embodiment of the present application;

[0060] Figure 8 It is a schematic diagram of the composition structure of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0062] Artificial intelligence technology is a comprehensive discipline involving a wide range of fields, including both hardware-level technologies and software-level technologies. The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0063] Among them, computer vision technology (CV) is a science that studies how to make machines "see". Further, it refers to using cameras and computers to replace human eyes to perform machine vision such as target recognition, tracking, and measurement on targets, and further perform graphic processing to make the computer process into images that are more suitable for human eyes to observe or be transmitted to instruments for detection. As a scientific discipline, computer vision studies related theories and technologies and attempts to establish an artificial intelligence system that can obtain information from images or multi-dimensional data. Computer vision technology usually includes technologies such as image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, and also includes common biometric recognition technologies such as face recognition and fingerprint recognition. The present application relates to image recognition technology in artificial intelligence. Using image recognition technology, the transformation function of an application program can be detected to determine whether the transformation function is normal, thereby improving the accuracy of data verification.

[0064] Please refer to Figure 1 , Figure 1 It is a network architecture diagram of a data verification system provided by an embodiment of the present application, as Figure 1As shown, the computer device can interact with user terminals. The number of user terminals can be one or more. When the number of user terminals is more than one, the user terminals can include Figure 1 102a and 102b in Figure 1 and the computer device can be 101 in

[0065] Taking the user terminal 102a as an example, the computer device 101 can identify the initial marker point corresponding to the target object in the multimedia data targeted by the transformation trigger instruction of the transformation function in response to the transformation trigger instruction of the transformation function from the user terminal 102a. Further, the computer device 101 performs transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain the updated multimedia data. The transformation trigger instruction includes the expected transformation value of the transformation processing. The computer device 101 identifies the updated marker point corresponding to the initial marker point in the updated multimedia data, and obtains the marker difference value between the initial marker point and the updated marker point, so as to determine the detection result of the transformation function according to the marker difference value.

[0066] It can be understood that the computer devices mentioned in the embodiments of the present application include, but are not limited to, terminal devices or servers. In other words, the computer device or user device can be a server or a terminal device, or a system composed of a server and a terminal device. Among them, the above-mentioned terminal device can be an electronic device, including but not limited to mobile phones, tablet computers, desktop computers, laptop computers, palmtop computers, augmented reality / virtual reality (AR / VR) devices, head-mounted displays, wearable devices, smart speakers, digital cameras, cameras, and other mobile internet devices (MIDs) with network access capabilities. Among them, the user terminal has a display function. Among them, the above-mentioned server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0067] Optionally, the data involved in the embodiments of the present application can be stored in a server, in the memory of a computer device, or can be stored based on cloud storage technology, without limitation here.

[0068] Further, please refer to Figure 2 , Figure 2 which is a schematic diagram of an application scenario of a data verification method provided by an embodiment of the present application. As Figure 2As shown, the computer device 200 responds to a transformation trigger instruction for a transformation function, and identifies an initial marker point 2012 corresponding to a target object 2011 in the multimedia data 201 targeted by the transformation trigger instruction for the transformation function. Among them, the transformation trigger instruction includes an expected transformation value for the transformation process. For example, the expected transformation value can be 10%. The computer device 200 performs a transformation process on the target object 201 in the multimedia data 201 based on this transformation trigger instruction to obtain updated multimedia data 202. The computer device 200 identifies an updated marker point 2022 corresponding to the initial marker point 2012 in the updated multimedia data 202, and obtains a marker difference value between the initial marker point 2012 and the updated marker point 2022. For example, if the initial marker point is the pixel point corresponding to the shoulder of the target object in the multimedia data, the computer device can obtain the distance L1 between two initial marker points corresponding to the shoulder of the target object in the multimedia data, and obtain the distance L2 between two updated marker points corresponding to the shoulder of the target object in the updated multimedia data, then the difference between L1 and L2 can be determined as the marker difference value. The computer device 200 determines the detection result of this transformation function according to this marker difference value and the expected transformation value of the transformation process, where the detection result of the transformation function includes a valid transformation function result or an invalid transformation function result. For example, if the computer device determines that the difference data between the marker difference value and the expected transformation value of the transformation process belongs to the valid error range, it determines that the detection result of the transformation function is a valid transformation function result; if the computer device determines that the difference data between the marker difference value and the expected transformation value of the transformation process does not belong to the valid error range, it determines that the detection result of the transformation function is an invalid transformation function result, thereby realizing the detection of the transformation function.

[0069] Further, please refer to Figure 3 , Figure 3 which is a schematic flowchart of a data verification method provided by an embodiment of the present application; as Figure 3 shown, the method includes:

[0070] S101, in response to a transformation trigger instruction for a transformation function, identify an initial marker point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction.

[0071] In the embodiments of the present application, when a user sends a transformation trigger instruction for a transformation function through a user terminal, the computer device responds to the transformation trigger instruction for the transformation function and identifies the initial marker points corresponding to the target object in the multimedia data targeted by the transformation trigger instruction. The multimedia data may be video data or a single-frame picture, etc., which is not limited herein. If the multimedia data is video data, the processing of steps S101 to S104 will be performed for each frame of the video data. The target object may refer to the object indicated by the image in the multimedia data. For example, the target object may refer to a person, or may refer to an animal or a plant, etc. For example, the target object may be the person shown as 2011 in Figure 2 and 2031 in Figure 2 . The initial marker points are used to mark the pixel points corresponding to the key parts of the target object in the multimedia data. For example, when the target object is a person, the key parts of the target object may refer to the shoulders, waist, legs, hips or other key parts of the target object. The initial marker points can be used to mark the pixel points corresponding to the shoulders, waist, legs, hips or other key parts of the target object in the multimedia data.

[0072] Optionally, the computer device may identify the initial marker points corresponding to the target object in the multimedia data targeted by the transformation trigger instruction based on a target detection model. Specifically, the computer device identifies the multimedia data targeted by the transformation trigger instruction based on the target detection model to determine the target object in the multimedia data; obtains the object features of the target object in the multimedia data, and determines the initial marker points corresponding to the target object from the marker points corresponding to the object features. The target detection model may include, but is not limited to, a model composed of one or more of networks such as Convolutional Neural Networks (CNN), Visual Geometry Group (VGG), or Residual Network (ResNet).

[0073] In a specific implementation, the computer device can identify the multimedia data targeted by the transformation trigger instruction based on the object detection model to determine the target object in the multimedia data; obtain the features of the pixel points corresponding to the target object, and determine the object features corresponding to the key parts of the target object based on the structure of the target object and the relative positions between the key parts of the target object; identify the boundary points of the key parts of the target object, and determine the boundary points of the key parts of the target object as the initial marking points corresponding to the target object. For example, assume the target object is a person. Then the structure of the person can be a structure composed of the person's facial features, limbs, and torso, etc. The person's facial features are above the limbs, and the person's limbs are on the left or right side of the torso. Assume the key parts of the person include the legs and shoulders, etc. Determine the person's legs and shoulders based on the relative positions between the person's legs and shoulders. For example, the computer device can identify the person in the multimedia data based on the object detection model, obtain the features of the pixel points corresponding to the person, and perform feature extraction on the features of the pixel points corresponding to the person based on the structure of the person (structures such as facial features, limbs, and torso) and the relative positions between the key parts of the person to determine the object features corresponding to the key parts of the person, that is, the object features corresponding to the shoulders and the object features corresponding to the legs, etc. The initial marking points corresponding to the target object can be determined from the marking points of the object features corresponding to the key parts in the multimedia data. The initial marking points can represent the marking points corresponding to the person's shoulders and legs. Taking the shoulders as an example, identify the features indicating the boundary points of the shoulders in the object features corresponding to the shoulders, that is, the features of the boundary points corresponding to the left shoulder and the features of the boundary points corresponding to the right shoulder. The distance between the boundary points corresponding to the left shoulder and the boundary points corresponding to the right shoulder represents the maximum shoulder width of the person. The computer device can determine the pixel points corresponding to the features of the boundary points corresponding to the left shoulder and the features of the boundary points corresponding to the right shoulder in the multimedia data as the initial marking points corresponding to the target object.

[0074] Optionally, the computer device can determine the initial marking points corresponding to the target object in the multimedia data based on the marking parameters in the multimedia data. Specifically, the computer device obtains the target marking parameters associated with the transformation function corresponding to the transformation trigger instruction, and obtains the target marking parameter range corresponding to the target marking parameters; in the multimedia data targeted by the transformation trigger instruction, determine the pixel points whose pixel parameters belong to the target marking parameter range as the initial marking points corresponding to the target object in the multimedia data; the initial marking points with the same target marking parameters are the initial marking point pairs. The pixel parameters belonging to the target marking parameter range can be considered as the target marking parameters.

[0075] In a specific implementation, the computer device may first obtain the target marker parameters associated with the transformation function corresponding to the transformation trigger instruction. For example, the target marker parameters associated with the transformation function refer to the marker parameters corresponding to a certain key part (such as the shoulder) (that is, the transformation function is a transformation process for the shoulder of the target object, and no transformation process is performed on other key parts of the target object such as the waist and legs). Then, the computer device obtains the range of the target marker parameters corresponding to the shoulder of the target object. By detecting the pixel points in the multimedia data, if a pixel point whose pixel parameter belongs to the range of the target marker parameters is detected, then the pixel point whose pixel parameter belongs to the range of the target marker parameters is determined as the initial marker point corresponding to the target object in the multimedia data.

[0076] Among them, the marker parameter may refer to the pixel value of the pixel point in the multimedia data. For example, the red (Red, R) value, green (Green, G) value, and blue (Blue, B) value of the pixel point in the multimedia data, that is, the RGB value of the pixel point. For example, the computer device obtains at least two marker parameters, including red, green, and blue, and the target key part (that is, the key part for which the transformation process is performed based on the transformation function) is marked based on red, that is, the range of the target marker parameters is red. Then, the marker parameter corresponding to "red" obtained from the at least two marker parameters is used as the target marker parameter. Specifically, the range of the marker parameter may refer to the parameter range corresponding to the marker parameter. For example, the marker parameter includes red, green, and blue. Among them, the range of the marker parameter "red" may be R>250, G<5, and B<5; the range of the marker parameter "green" may be R<5, G>250, and B<5; the range of the marker parameter "blue" may be R<5, G<5, and B>250. Assume that the target key part to be detected is marked based on "red", then the range of the target marker parameters is determined as the range of the marker parameters corresponding to "red", that is, "R>250, G<5, and B<5". The pixel points corresponding to the marker parameters that satisfy this range of the target marker parameters are determined as the initial marker points corresponding to the target object, and the pixel points corresponding to the marker parameters belonging to the same range of the target marker parameters are determined as the initial marker point pairs.

[0077] For example, a computer device obtains at least two marker parameters in multimedia data. For example, marker parameter 1, marker parameter 2, marker parameter 3, marker parameter 4, marker parameter 5, and marker parameter 6 are obtained. The six marker parameters are respectively (250, 0, 0), (250, 0, 0), (0, 250, 0), (0, 250, 0), (125, 0, 0), and (200, 0, 0). The target marker parameter range includes (R > 250, G < 5, and B < 5). It can be seen that the marker parameters among the six marker parameters that belong to the target marker parameter range are marker parameter 1 and marker parameter 2. Then, the pixel points corresponding to marker parameter 1 and marker parameter 2 in the multimedia data are determined as the initial marker points corresponding to the target object, and the initial marker points with the same marker parameters are determined as the initial marker point pairs, that is, marker parameter 1 and marker parameter 2 are the initial marker point pairs.

[0078] Among them, when the computer device obtains the initial marker points corresponding to the target object in the multimedia data, it detects at least two pixel points in the multimedia data; if the pixel parameters corresponding to the h-th pixel point among the at least two pixel points belong to the target marker parameter range, the h-th pixel point is determined as the initial marker point corresponding to the target object in the multimedia data; h is a positive integer; if the pixel parameters corresponding to the h-th pixel point among the at least two pixel points do not belong to the target marker parameter range, the pixel parameters corresponding to the (h + 1)-th pixel point are detected, and it is detected whether the pixel parameters corresponding to the (h + 1)-th pixel point belong to the target marker parameter range to detect whether the (h + 1)-th pixel point is the initial marker point corresponding to the target object in the multimedia data; until the detection of all pixel points in the multimedia data is completed, the initial marker points corresponding to the target object in the multimedia data are obtained, and the initial marker points with the same target marker parameters are determined as the initial marker point pairs.

[0079] In a specific implementation, the computer device detects at least two pixel points in the multimedia data. For example, it can scan the multimedia data column by column from left to right. When it detects that the pixel parameters corresponding to a pixel point (i.e., the h-th pixel point) in a certain column belong to the target marking parameter range, it determines this pixel point (i.e., the h-th pixel point) as the initial marking point corresponding to the target object in the multimedia data. Further, the computer device can continue to scan the multimedia data column by column from right to left. When it detects that the pixel parameters corresponding to a pixel point in a certain column belong to the target marking parameter range, the computer device determines this pixel point as the initial marking point corresponding to the target object in the multimedia data. Optionally, the computer device can also directly scan the multimedia data column by column (such as scanning column by column from left to right or from right to left, etc.); or it can also directly scan the multimedia data row by row. Among them, since there may be a symmetry relationship in the key parts of the target object, at this time, it can first scan column by column from left to right. After detecting that the pixel parameters belong to the target marking parameter range (i.e., the initial marking point), then scan column by column from right to left to improve the recognition efficiency of the key parts. For example, the shoulders of a person (there are left and right shoulders), the waist (there are two sides), or the legs (there are two legs), etc.

[0080] Optionally, the computer device can obtain the target marking parameter range and the number of detection marking points corresponding to the target key parts to be detected. Then it scans the multimedia data from left to right. After detecting a pixel point in the multimedia data whose pixel parameters belong to the target marking parameter range, it determines this pixel point whose pixel parameters belong to the target marking parameter range as the initial marking point. Then it scans the multimedia data from right to left. If it detects a pixel point whose pixel parameters belong to the target marking parameter range, it determines the pixel point whose pixel parameters belong to the target marking parameter range in this scan as the initial marking point. When the number of obtained initial marking points is equal to the number of detection marking points, the computer device can end the recognition of the multimedia data, thereby improving the data recognition efficiency. For example, when the number of detection marking points corresponding to the target key parts is 2, the computer device scans the multimedia data from left to right. If it detects an initial marking point in the multimedia data whose pixel parameters belong to the target marking parameter range, then it scans the multimedia data from right to left. If it detects another initial marking point in the multimedia data whose pixel parameters belong to the target marking parameter range, it can be known that the number of initial marking points at this time is 2 and is equal to the number of detection marking points. Then the computer device can end the recognition of the multimedia data, thereby improving the data recognition efficiency.

[0081] Optionally, the marking parameter may also refer to a parameter for marking pixels corresponding to key parts of a target object in multimedia data. For example, different geometric shapes (such as triangles, rectangles, or circles, etc.) or character markings can be used to mark the pixels corresponding to key parts of the target object in multimedia data. As Figure 4a shown, Figure 4a FIG. Figure 4a is a schematic diagram for determining a marking parameter provided by an embodiment of the present application. For example, the multimedia data includes marking parameters such as triangles, rectangles, and circles for marking a target object, and the target marking parameter range includes triangles. The computer device identifies geometric shapes such as triangles, rectangles, and circles used to mark the target object in the multimedia data, and determines the marking parameter that matches the target marking parameter range among the geometric shapes of triangles, rectangles, and circles as the target marking parameter. That is, the computer device determines the triangle belonging to the target marking parameter range as the target marking parameter, and then the computer device determines the pixels corresponding to the triangle in the multimedia data as the initial marking points corresponding to the target object. By obtaining the pixels corresponding to the geometric shapes belonging to the target marking parameter range in the multimedia data and determining the pixels as the initial marking points corresponding to the target object, the pixels corresponding to all key parts (such as legs and waist, etc.) of the target object in the multimedia data can be obtained, that is, all the initial marking points of the target object.

[0082] Optionally, the computer device can also determine all the marking parameters in the multimedia data. The marking parameter can refer to different key parts, for example, including the shoulder of the target object, the waist of the target object, and the legs of the target object, etc. Since the transformation trigger instruction of the transformation function may be for transformation processing of one or at least two key parts, the computer device identifies all the marking parameters in the multimedia data, and then determines the key parts (target key parts) to be detected from all the marking parameters. Specifically, the marking parameter that belongs to the target marking parameter range among at least two marking parameters is determined as the target marking parameter. The target marking parameter range is used to represent the target marking parameter corresponding to the target key part, that is, the target marking parameter range is used to screen the target marking parameter corresponding to the target key part to be detected from at least two marking parameters. Since the computer device determines the target marking parameter from multiple marking parameters in the above process, the pixels corresponding to the target marking parameter in the multimedia data can be directly determined as the initial marking points corresponding to the target object.

[0083] Wherein, when the marking parameter is a color, a color with a large color difference from the color existing in the multimedia data can be used to mark the key parts of the target object, and the marking parameter range is determined according to the color with a large color difference from the color existing in the multimedia data, so that when detecting the marking parameter in the multimedia data, since the marking parameter has a large color difference from the color existing in the multimedia data, the recognition is relatively simple, the probability of misrecognition is reduced, and the recognition efficiency is improved. Specifically, the computer device can map the pixel values of the pixel points in the multimedia data to a color map, and determine the marking parameter range based on the color map. Specifically, the computer device obtains the pixel values of at least two pixel points in the multimedia data, maps the at least two pixel points to the color map based on the pixel values of the at least two pixel points; performs clustering processing on the at least two pixel points based on the positions of the at least two pixel points in the color map to obtain a multimedia color cluster; determines the marking parameter range according to the map color values in the color map whose color distances from the multimedia color cluster are greater than the color difference threshold, wherein the marking parameter range includes the target marking parameter range. Optionally, if the number of multimedia color clusters is z, the computer device determines the marking parameter range according to the map color values in the color map whose color distances from each of the z multimedia color clusters are greater than the color difference threshold, and z is a positive integer.

[0084] As Figure 4b shown, Figure 4bThis is a schematic diagram of a scenario for determining the range of marker parameters provided by an embodiment of the present application. The multimedia data 41 includes 80 pixel points, and each pixel point corresponds to a pixel value. The computer device obtains the pixel value of each pixel point and maps the 80 pixel points to the color map 42 based on the pixel value of each pixel point. Among them, the colors in the color map 42 change gradually pixel by pixel. That is, the closer the color distance between two pixel points in the color map, the closer the pixel values of the pixel points, that is, the more similar the colors; the farther the color distance between two pixel points in the color map, the greater the difference in the pixel values of these two pixel points, that is, the greater the color difference. The computer device can perform clustering processing on the 80 pixel points based on their positions in the color map to obtain a multimedia color cluster. For example, 3 multimedia color clusters are obtained. According to the map color values in the color map whose color distances from the 3 color clusters are greater than the color difference threshold, the range of the marker parameters is determined. As shown in the color map 43, the 3 solid circles in the color map respectively represent a multimedia color cluster. Based on the multimedia color cluster, the color distance corresponding to the color difference threshold is extended outward to obtain a color similarity range, that is, the 3 dashed circles in the color map 43. The color distance between the map pixel points within the dashed circle and the pixel values in the corresponding multimedia color cluster is less than or equal to the color difference threshold, and the color distance between the map color values corresponding to the map pixel points outside the 3 dashed circles and the multimedia color cluster is greater than the color difference threshold. That is, the map color values corresponding to the map pixel points not within any of the 3 dashed circles can be used to form the range of the marker parameters. Specifically, the range outside the 3 dashed circles can be determined as the candidate area, and a marker area is randomly selected from the candidate area, and the map color values corresponding to the map pixel points in the marker area are used to form the range of the marker parameters. Further, the number of types of key parts can be obtained, and the marker area is selected based on the number of types of the key parts, that is, one key part corresponds to one marker area. For example, if the pixel point t1 is within the multimedia color cluster, the color distance between the map color value of the pixel point t1 and the multimedia color cluster is less than the color difference threshold; while the pixel point t2 is in the candidate area outside the multimedia color cluster, the color distance between the map color value of the pixel point t2 and the multimedia color cluster is greater than the color difference threshold. The computer device can determine the range to which the map color value corresponding to the pixel point t2 belongs as the range of the marker parameters, and the range of the marker parameters includes the target range of the marker parameters.

[0085] S102. Perform transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain the updated multimedia data.

[0086] In the embodiments of the present application, based on a transformation trigger instruction, transformation processing is performed on a target object in multimedia data to obtain updated multimedia data, that is, the target object in the updated multimedia data is the target object after the transformation processing. The transformation trigger instruction includes an expected transformation value, and the expected transformation value can refer to the amount of data to be transformed or the percentage of transformation, etc. Among them, the transformation function is used to transform the key parts of the target object, so as to realize the transformation of the key parts of the target object. For example, the waist width of the target object is reduced, the leg length is increased, the leg width is reduced, the shoulder width is reduced, etc., so as to realize the beautification of the key parts of the target object, and further realize the beautification of the target object.

[0087] Optionally, the transformation function can perform transformations on all the key parts corresponding to the target object; or, the transformation function can perform transformations on one or at least two of the key parts corresponding to the target object. When the transformation function performs transformations on all the key parts corresponding to the target object, the computer device responds to the transformation trigger instruction for the transformation function and performs transformation processing on all the key parts of the target object in the multimedia data. For example, it includes performing transformations on the shoulder width of the target object, the leg length of the target object, and the waist width of the target object, etc. When the transformation function performs transformations on one or at least two of the key parts corresponding to the target object, the computer device can implement transformation processing on the one or at least two key parts of the target object according to the transformation trigger instruction of the user terminal for a specific one or at least two key parts. For example, if the transformation trigger instruction of the user terminal is the long-leg operation plus the narrow-shoulder operation in the transformation function, the computer device responds to this transformation trigger instruction and performs transformation processing on the leg length and shoulder width of the target object, and does not perform transformation processing on other key parts of the target object such as the waist width. Or, if the trigger operation of the user terminal is the long-leg operation in the transformation function, the computer device responds to this transformation trigger instruction and performs transformation processing on the legs of the target object, and does not perform transformation processing on other key parts of the target object, such as the waist or shoulders, etc. As Figure 5 shown, Figure 5It is a schematic diagram for transforming a target object provided by an embodiment of the present application. Among them, the transformation trigger instruction of the transformation function is a transformation for a long leg operation, and the expected transformation value of the transformation trigger instruction is 20%. The computer device responds to this transformation trigger instruction and obtains the target parameter type corresponding to the transformation trigger instruction of the transformation function. The computer device obtains the parameter type of the corresponding initial marking point of the target object, and determines the initial marking point whose parameter type matches the target parameter type as the target initial marking point. For example, the target parameter type is the parameter type corresponding to the leg of the target object. Assume that the parameter type corresponding to the leg of the target object is (R>250, G<5, and B<5), that is, red. Then the computer device uses the initial marking points whose parameter types belong to (R>250, G<5, and B<5) as the target initial marking points. For example, if the computer device determines that the target initial marking points in the multimedia data 501 are pixel point 1 and pixel point 2, then the computer device performs a transformation process on the leg of the target object based on these two target initial marking points, stretches the leg by 20%, and obtains the leg feature 502 of the target object after the transformation process. Based on this leg feature 502, the updated multimedia data 503 is obtained, so that Figure 5 the leg length of the target object in the updated multimedia data 503 is greater than the leg length of the target object in the multimedia data 501.

[0088] S103. Identify the updated marking points corresponding to the initial marking points in the updated multimedia data.

[0089] In the embodiment of the present application, the computer device can identify the updated multimedia data based on the target detection model, determine the target object in the updated multimedia data; obtain the object feature of the target object in the updated multimedia data, and determine the updated marking point corresponding to the initial marking point from the marking points corresponding to the object feature of the target object in the updated multimedia data. The specific method for identifying the updated marking points corresponding to the initial marking points in the updated multimedia data based on the target detection model can refer to the method for identifying the initial marking points corresponding to the target object in the multimedia data in step S101, which will not be elaborated here.

[0090] Optionally, the computer device can also identify the updated marker points corresponding to the initial marker points in the updated multimedia data based on the updated marker parameter in the updated multimedia data. Specifically, the computer device obtains the target updated marker parameter associated with the transformation function corresponding to the transformation trigger instruction, obtains the target updated marker parameter range corresponding to the target updated marker parameter, and in the updated multimedia data, determines the pixel points whose updated pixel parameters belong to the target updated marker parameter range as the updated marker points corresponding to the initial marker points in the updated multimedia data. The updated marker points with the same target updated marker parameter are the updated marker point pairs. Among them, the updated marker parameter can refer to the pixel value of the pixel point in the updated multimedia data. For example, the R value, G value, and B value of the pixel point in the updated multimedia data. For example, the computer device obtains at least two updated marker parameters, including red, green, and blue, and the target key part (such as the shoulder of a person) in the updated multimedia data is marked based on red, that is, the target marker parameter range is red, then the marker parameter corresponding to "red" obtained from the at least two updated marker parameters is used as the marker parameter. Specifically, the updated marker parameter range can refer to the parameter range corresponding to the updated marker parameter. For example, the updated marker parameters include red, green, and blue. Among them, the parameter range corresponding to the updated marker parameter "red" can be R>250, G<5, and B<5; the parameter range corresponding to the updated marker parameter "green" can be R<5, G>250, and B<5; the parameter range corresponding to the updated marker parameter "blue" can be R<5, G<5, and B>250. The specific method for identifying the updated marker points corresponding to the initial marker points in the updated multimedia data can refer to the description of the identification process of the initial marker points in step S101, which will not be elaborated here.

[0091] S104. Obtain the marker difference value between the initial marker point and the updated marker point.

[0092] In the embodiment of the present application, the number of initial marker points is at least two, and the number of updated marker points is at least two. The computer device can obtain the initial distance between the initial marker points with the same parameter type among the at least two initial marker points, and the distance between the updated marker points with the same parameter type among the at least two updated marker points, so as to determine the marker difference value between the initial marker point and the updated marker point according to the difference between the initial distance and the updated distance.

[0093] Specifically, the computer device obtains the initial parameters corresponding to each of at least two initial marker points, obtains the parameter types of the initial parameters corresponding to each of the initial marker points, and performs grouping processing on the at least two initial marker points based on the parameter types of the initial parameters to obtain m pairs of initial marker points. Among them, the m pairs of initial marker points include the initial marker point pair i, the initial marker point pair i includes a first initial marker point and a second initial marker point, m is a positive integer, and i is a positive integer. Among them, the initial parameter refers to the marker parameter corresponding to the initial marker point in the multimedia data. Since the initial marker points corresponding to the target object in the multimedia data include the initial marker points corresponding to the shoulders of the target object, the initial marker points corresponding to the waist of the target object, and the initial marker points corresponding to the legs of the target object, and the parameter types of the initial parameters corresponding to the three types of initial marker points are different. For example, the parameter type of the initial parameter corresponding to the initial marker point may refer to the pixel value of the initial marker point. For example, the initial marker point whose pixel value of the initial marker point belongs to the range of R>250, G<5, and B<5 (i.e., the red initial marker point) is one parameter type, the initial marker point whose pixel value of the initial marker point belongs to the range of R<5, G>250, and B<5 (i.e., the green initial marker point) is another parameter type, and the initial marker point whose pixel value of the initial marker point belongs to the range of R<5, G<5, and B>250 (i.e., the blue initial marker point) is yet another parameter type.

[0094] That is to say, the computer device can determine whether each initial marker point is of the same parameter type by obtaining the pixel values of the respective initial marker points corresponding to the target object in the multimedia data, group two initial marker points of the same type to obtain pairs of initial marker points, and thus obtain m pairs of initial marker points. For each key part of the target object, there is a corresponding pair of initial marker points. For example, the shoulders of the target object correspond to a pair of initial marker points (i.e., two initial marker points), the waist of the target object corresponds to another pair of initial marker points, and the legs of the target object correspond to yet another pair of initial marker points. The initial marker point pair i may refer to the pair of initial marker points corresponding to the shoulders of the target object, and the pair of initial marker points corresponding to the shoulders of the target object includes a first initial marker point (such as the left shoulder) and a second initial marker point (such as the right shoulder). Optionally, the initial marker point pair i may refer to the pair of initial marker points corresponding to the waist of the target object, or the initial marker point pair i may refer to the pair of initial marker points corresponding to the legs of the target object, and so on. It can be understood that the embodiments of the present application process the initial marker points corresponding to any one key part of the target object, and the processing method for the initial marker points corresponding to other key parts of the target object can refer to the processing method for the initial marker points corresponding to this arbitrary key part, and will not be described in detail here.

[0095] Optionally, the initial parameter corresponding to the initial marking point may be a geometric figure used to mark the initial marking point in the multimedia data. Then, the parameter type of the initial parameter may be the type of geometric figure used to mark the initial marking point in the multimedia data. The type of geometric figure may include a triangle, a rectangle, a circle, etc. The computer device groups the initial marking points corresponding to the geometric figures of the same type by obtaining the type of geometric figure of the initial marking points in the multimedia data, and obtains m pairs of initial marking points.

[0096] Further, after determining the initial marking point pair i in the m pairs of initial marking points, the computer device obtains the update parameters respectively corresponding to each of the at least two updated marking points, obtains the parameter type of the update parameters respectively corresponding to each of the at least two updated marking points, and performs grouping processing on the at least two updated marking points based on the parameter type of the update parameters to obtain m pairs of updated marking points. Among them, the m pairs of updated marking points include the updated marking point pair i, and the updated marking point pair i includes a first updated marking point and a second updated marking point. The update parameter refers to the marking parameter corresponding to the updated marking point in the updated multimedia data. Among them, since the transformation function is used to transform the key parts of the target object, to realize the transformation of the key parts of the target object, such as making the waist width of the target object smaller, the leg length of the target object longer, the shoulder width of the target object smaller, etc., to realize the beautification of the target object, but not to make the number of key parts of the target object less. Therefore, the number of updated marking points corresponding to the updated target object is equal to the number of initial marking points before transformation.

[0097] Therefore, the updated marking points corresponding to the initial marking points in the updated multimedia data include the updated marking points corresponding to the shoulders of the target object, the updated marking points corresponding to the waist of the target object, and the updated marking points corresponding to the legs of the target object, and the parameter types of the update parameters corresponding to the three types of updated marking points are different. Among them, the updated marking points corresponding to the shoulders correspond to the initial marking points corresponding to the shoulders, the updated marking points corresponding to the waist correspond to the initial marking points corresponding to the waist, and the updated marking points corresponding to the legs correspond to the initial marking points corresponding to the legs, etc. For example, the parameter type of the update parameter corresponding to the updated marking point may refer to the pixel value of the updated marking point. For example, the updated marking points with pixel values of the updated marking point belonging to the range of R>250, G<5, and B<5 (i.e., red updated marking points) are of one parameter type, the updated marking points with pixel values of the updated marking point belonging to the range of R<5, G>250, and B<5 (i.e., green updated marking points) are of another parameter type, and the updated marking points with pixel values of the updated marking point belonging to the range of R<5, G<5, and B>250 (i.e., blue updated marking points) are of yet another parameter type.

[0098] That is to say, the computer device can determine whether each updated marker point corresponding to the target object in the updated multimedia data is of the same type by obtaining the pixel values of the updated marker points, group two updated marker points of the same type to obtain an updated marker point pair, and thus obtain m updated marker point pairs. For each key part of the target object, there is a corresponding updated marker point pair. For example, the shoulder of the target object corresponds to an updated marker point pair, the waist of the target object corresponds to another updated marker point pair, and the leg of the target object corresponds to yet another updated marker point pair. The updated marker point pair i may refer to the updated marker point pair corresponding to the shoulder of the target object. This updated marker point pair i corresponds to the initial marker point pair i, and the updated marker point pair corresponding to the shoulder of the target object includes a first updated marker point (such as the left shoulder) and a second updated marker point (such as the right shoulder). Optionally, the updated marker point pair i may refer to the updated marker point pair corresponding to the waist of the target object, or the updated marker point pair i may refer to the updated marker point pair corresponding to the leg of the target object, and so on. It should be noted that if the initial marker point i is the initial marker point corresponding to the shoulder of the target object, the updated marker point i is also the updated marker point corresponding to the shoulder of the target object; if the initial marker point i is the initial marker point corresponding to the waist of the target object, the updated marker point i is also the updated marker point corresponding to the waist of the target object; if the initial marker point i is the initial marker point corresponding to the leg of the target object, the updated marker point i is also the updated marker point corresponding to the leg of the target object; that is, it means that the key part of the target object corresponding to the initial marker point is the same as the key part corresponding to the updated marker point. The embodiments of the present application process the updated marker points corresponding to any one key part of the target object. The processing method for the updated marker points corresponding to other key parts of the target object can refer to the processing method for the updated marker points corresponding to this any one key part, and will not be described in detail here.

[0099] Optionally, if the parameter type of the update parameter corresponding to the updated marker point refers to the type of the geometric figure used to mark the updated marker point in the multimedia data, the computer device can also obtain the type of the geometric figure of the updated marker point in the multimedia data, group the updated marker points corresponding to the geometric figures of the same type, and obtain m updated marker point pairs.

[0100] Further, the computer device obtains the first initial pixel position of the first initial marker point in the multimedia data and the second initial pixel position of the second initial marker point in the multimedia data in each initial marker point pair, and determines the initial distance of the initial marker point pair i according to the first initial pixel position and the second initial pixel position.

[0101] Specifically, the computer device can determine a red initial marker point by obtaining the pixel value of each pixel point in the multimedia data and determining that the pixel point whose corresponding pixel value satisfies R>250, G<5, and B<5 is a red initial marker point, for exampleFigure 2 the marked points corresponding to the shoulders of the target object in the [multimedia data], the initial marker point pair i may include Figure 2 two initial marker points corresponding to the shoulders of the target object in the [multimedia data], for example, they may be denoted as the initial marker point a1 and the initial marker point a2. Then the first initial marker point may refer to the initial marker point a1, and the second initial marker point may refer to the initial marker point a2. The computer device may establish a rectangular coordinate system with any position of the [multimedia data] as the origin of coordinates. For example, a rectangular coordinate system is established with the upper left corner position of the [multimedia data] as the origin of coordinates. When the computer device obtains the red initial marker point a1 and the red initial marker point a2 in the [multimedia data], it may respectively obtain the position coordinates of the two red marker points a1 and a2, and by calculating the distance between the position coordinates of a1 and a2, obtain the distance value between a1 and a2, and obtain the initial distance of the initial marker point pair i, that is, the shoulder width of the target object in the [multimedia data].

[0102] In a specific implementation, the computer device may scan each column of pixel points from left to right starting from the leftmost column of the [multimedia data] (that is, the ordinate of the pixel points is 0). When there is a pixel point in a certain column of pixel points whose pixel value satisfies R>250, G<5, and B<5, then this pixel point is determined as the red initial marker point a1, and the abscissa of the red initial marker point a1 is recorded as x1. Correspondingly, the computer device may scan each column of pixel points from right to left starting from the rightmost column of the [multimedia data]. When there is a pixel point in a certain column of pixel points whose pixel value satisfies R>250, G<5, and B<5, then this pixel point is determined as the red initial marker point a2, and the abscissa of the red initial marker point a2 is recorded as x2, and the difference between x2 and x1 is determined as the distance value between a1 and a2.

[0103] It can be understood that for the initial marker points corresponding to other key parts of the target object, the distance between the initial marker points corresponding to other key parts can also be obtained by referring to the above method, so as to obtain the waist width, leg length, etc. of the target object. It should be noted that when determining the distance value between the initial marker points corresponding to the legs of the target object, the computer device may start scanning column by column from each row of pixel points at the top of the [multimedia data] to determine the position coordinates of the initial marker points corresponding to the legs of the target object, and start scanning column by column from each row of pixel points at the bottom of the [multimedia data] to determine the position coordinates of the initial marker points corresponding to the legs of the target object, so as to determine the leg length of the target object according to the distance between the position coordinates of the two initial marker points.

[0104] Further, the computer device obtains the first updated pixel position of the first updated marker point in the updated multimedia data for each updated marker point pair, and the second updated pixel position of the second updated marker point in the updated multimedia data. According to the first updated pixel position and the second updated pixel position, the computer device determines the updated distance of the updated marker point pair i. Among them, the computer device determining the updated distance of the updated marker point pair i may refer to the computer device determining the distance between two updated marker points corresponding to the shoulders of the target object in the updated multimedia data. The specific method for determining the updated distance of the updated marker point pair i may refer to the method for determining the initial distance of the initial marker point pair i described above, and will not be described in detail here.

[0105] Since the computer device obtains the initial distance of the initial marker point pair i in the multimedia data before the update, and the updated distance of the updated marker point pair i in the updated multimedia data, the computer device can determine the distance difference between the initial distance of the initial marker point pair i and the updated distance of the updated marker point pair i as the i-th marker difference value. That is to say, since the initial markers include the initial markers corresponding to the shoulders of the target object, the initial markers corresponding to the waist of the target object, and the initial markers corresponding to the legs of the target object, for any key part (shoulder, waist, or leg) of the target object, the above method can be used to calculate the initial distance corresponding to the key part of the target object, that is, the shoulder width of the target object in the original multimedia data, the waist width of the target object in the original multimedia data, and the leg length of the target object in the original multimedia data. Correspondingly, the above method can be used to calculate the updated distance corresponding to the key part of the target object, that is, the shoulder width of the target object in the updated multimedia data, the waist width of the target object in the updated multimedia data, and the leg length of the target object in the updated multimedia data. Optionally, the computer device can also directly identify the key parts such as the waist, shoulder, or leg of the target object in the multimedia data based on the object detection model, so as to determine the waist width, shoulder width, and leg length of the target object in the multimedia data based on the object detection model, etc., and thus obtain the initial distance of the initial marker point pair i. And, the computer device can also directly identify the key parts such as the waist, shoulder, or leg of the target object in the updated multimedia data based on the object detection model, so as to determine the waist width, shoulder width, and leg length of the target object in the updated multimedia data based on the object detection model, etc., and thus obtain the updated distance of the updated marker point pair i.

[0106] Optionally, the method for the computer device to determine the distance difference between the initial distance of the initial marker point pair i and the updated distance of the updated marker point pair i as the i-th marker difference value may include: The computer device obtains the transformed distance of the transformed marker point pair i of the target object in the target multimedia data by performing transformation processing on the multimedia data when the expected transformation value corresponding to the transformation trigger instruction is the largest; determines the i-th marker difference value according to the initial distance, the updated distance, and the transformed distance. The computer device may calculate the i-th marker difference value in the manner of formula (1-1):

[0107]

[0108] where q is the i-th marker difference value, and x 0％ is the initial distance of the initial marker point pair i, and x c is the updated distance of the updated marker point pair i, and x 100％ is the distance between the transformed marker point pairs i of the target object in the target multimedia data when the expected transformation value is the largest. The computer device can calculate the marker difference value corresponding to any key part of the target object through formula (1-1). For example, if the marker difference value corresponding to the shoulder of the target object is q1, the marker difference value corresponding to the waist of the target object is q2, and the marker difference value corresponding to the leg of the target object is q3, then any one of the values q1, q2, and q3 can be determined as the marker difference value between the initial marker point and the updated marker point. Optionally, the computer device may also determine all of q1, q2, and q3 as the marker difference value between the initial marker point and the updated marker point.

[0109] S105. Determine the detection result of the transformation function according to the marker difference value and the expected transformation value.

[0110] In the embodiments of the present application, since the computer device obtains the initial marker points corresponding to the target object in the multimedia data and the updated marker points corresponding to the initial marker points in the updated multimedia data through the above steps, the computer device can obtain the marker difference value between the initial marker point and the updated marker point, and determine the detection result of the transformation function according to the marker difference value and the expected transformation value, so as to determine whether the transformation function is an effective transformation function or an ineffective transformation function, and detect whether the transformation function takes effect normally based on the detection result. Specifically, when the transformation function is an effective transformation function, it is determined that the transformation function can take effect normally; when the transformation function is an ineffective transformation function, it is determined that the transformation function cannot take effect normally.

[0111] Optionally, the method for the computer device to determine the detection result of the transformation function based on the marked difference value and the expected transformation value can be as follows: Obtain the difference data between the marked difference value and the expected transformation value; if the difference data belongs to the effective error range, determine that the detection result of the transformation function is a valid transformation function result, that is, determine that the transformation function is a valid transformation function and can take effect normally (i.e., work properly); if the difference data does not belong to the effective error range, determine that the detection result of the transformation function is an invalid transformation function result, that is, determine that the transformation function is an invalid transformation function and the transformation function cannot take effect normally (i.e., cannot work properly).

[0112] Specifically, the computer device can obtain the absolute difference between the marked difference value and the expected transformation value, and determine this absolute difference as the difference data between the marked difference value and the expected transformation value. If the absolute difference between the marked difference value and the expected transformation value belongs to the effective error range, determine that the detection result of the transformation function is a valid transformation function result; if the absolute difference between the marked difference value and the expected transformation value does not belong to the effective error range, determine that the detection result of the transformation function is an invalid transformation function result. Among them, the effective error value can be 2%, 3%, 4% or other values, and the corresponding effective error range is the range greater than zero and less than the effective error value. For example, if the effective error value is 2%, the effective error range is 0 < d < 2%, and d represents the effective error value. Optionally, the computer device can determine the detection result of the transformation function according to the formula (1-2):

[0113] |p - q| ≤ d (1-2)

[0114] where q is the marked difference value, p is the expected transformation value, and d is the effective error value.

[0115] In an embodiment of the present application, in response to a transformation trigger instruction for a transformation function, an initial marker point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction is identified; the target object in the multimedia data is subjected to transformation processing based on the transformation trigger instruction to obtain updated multimedia data; wherein, the transformation trigger instruction includes an expected transformation value; an updated marker point corresponding to the initial marker point in the updated multimedia data is identified; a marker difference value between the initial marker point and the updated marker point is obtained, and based on the marker difference value and the expected transformation value, a detection result of the transformation function is determined. Since the expected transformation value corresponding to the transformation trigger instruction can represent the theoretical transformation data volume by which the transformation trigger instruction for the transformation function can transform the multimedia data, and the marker difference value can represent the actual changed data volume of the multimedia data after the transformation processing, by comparing the theoretical transformation data volume and the actual changed data volume of the multimedia data, the detection result of the transformation function can be determined, that is, it can be determined whether the transformation function is effective or ineffective; since the marker points in the multimedia data before and after the transformation are identified, and the marker difference value between the initial marker point and the updated marker point, as well as the difference between the marker difference value and the expected transformation value, are compared, the accuracy of data verification can be improved.

[0116] Optionally, the computer device can identify all the marker points in the multimedia data and the updated multimedia data, and then screen all the marker points based on the transformation function to obtain target marker points, so as to determine the marker difference value based on the target marker points. Specifically, the computer device can identify the initial marker point corresponding to the target object in the multimedia data targeted by the transformation trigger instruction, and the marker parameter corresponding to the initial marker point belongs to the marker parameter range, that is, as long as the pixel points marked in the multimedia data can be detected; the computer device identifies the updated marker point corresponding to the initial marker point in the updated multimedia data, and the marker parameter corresponding to the updated marker point belongs to the marker parameter range, that is, as long as the pixel points marked in the updated multimedia data can be detected. Further, please refer to Figure 6 , Figure 6 is a schematic flowchart of a data verification method provided by an embodiment of the present application; as Figure 6 shown, the method includes:

[0117] S201, in response to a transformation trigger instruction for a transformation function, identify an initial marker point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction.

[0118] In an embodiment of the present application, in response to a transformation trigger instruction for a transformation function, a computer device identifies all the marked points corresponding to a target object in the multimedia data targeted by the transformation trigger instruction, and records all the identified marked points as initial marked points. Specifically, the computer device determines a pixel point in the multimedia data whose pixel parameters belong to a marked parameter range as an initial marked point, where the initial marked point refers to all the marked points existing in the multimedia data, that is, it includes the marked points corresponding to the key parts in the multimedia data.

[0119] S202. Perform a transformation process on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data.

[0120] S203. Identify updated marked points corresponding to the initial marked points in the updated multimedia data.

[0121] In an embodiment of the present application, the computer device identifies all the marked points corresponding to the target object in the updated multimedia data, and records all the identified marked points as updated marked points. Specifically, the computer device determines an updated pixel point in the updated multimedia data whose pixel parameters belong to an updated marked parameter range as an updated marked point, where the updated marked point refers to all the marked points existing in the updated multimedia data, that is, it includes the marked points corresponding to the key parts in the updated multimedia data. Among them, the specific implementation manners of steps S201 to S203 can refer to Figure 3 the descriptions of steps S101 to S103 in the corresponding embodiments, which will not be elaborated here.

[0122] S204. Obtain a marked difference value between the initial marked points and the updated marked points.

[0123] In an embodiment of the present application, if the number of initial marked points is at least two and the number of updated marked points is at least two; then the computer device can obtain the marked difference value between the initial marked points and the updated marked points through the following method:

[0124] First, the computer device obtains the target parameter type associated with the transformation function corresponding to the transformation trigger instruction, obtains the parameter type of each of the at least two initial marked points, and determines the initial marked points that match the target parameter type among the at least two initial marked points as target initial marked points. Among them, the target initial marked points include target initial marked point j and target initial marked point k, the parameter types of target initial marked point j and target initial marked point k are the same, j is a positive integer, and k is a positive integer. The target parameter type corresponding to the transformation function is used to indicate a specific key part of the target object, and the parameter type of the initial marked point is used to indicate which key part of the target object the initial marked point corresponds to.

[0125] Specifically, the computer device obtains the target parameter type associated with the transformation function corresponding to the transformation trigger instruction. For example, the target parameter type is the parameter type corresponding to the shoulder of the target object. Assume that the parameter type corresponding to the shoulder of the target object is (R > 250, G < 5, and B < 5), that is, red. Then the computer device obtains the parameter type of each of at least two initial marker points, that is, obtains the pixel value of each initial marker point, and determines the initial marker points whose pixel values belong to the range of (R > 250, G < 5, and B < 5) as the target initial marker points.

[0126] Second, the computer device obtains the third initial pixel position of the target initial marker point j in the multimedia data and the fourth initial pixel position of the target initial marker point k in the multimedia data, and determines the target initial distance between the target initial marker point j and the target initial marker point k according to the third initial pixel position and the fourth initial pixel position.

[0127] Specifically, the computer device can establish a rectangular coordinate system with any position of the multimedia data as the coordinate origin. For example, establish a rectangular coordinate system with the upper left corner position of the multimedia data as the coordinate origin. The computer device obtains the position coordinates of the target initial marker point j in this coordinate system to obtain the third initial pixel position, and further, the computer device obtains the position coordinates of the target initial marker point k in this coordinate system to obtain the fourth initial pixel position. Furthermore, the computer device can obtain the distance value between the position coordinates corresponding to the third initial pixel position and the position coordinates corresponding to the fourth initial pixel position, for example, it can be the distance value between the abscissas of the two coordinates, to obtain the target initial distance between the target initial marker point j and the target initial marker point k.

[0128] Third, the computer device obtains the parameter type of each of at least two updated marker points, and determines the updated marker points that match the target parameter type among the at least two updated marker points as the target updated marker points. Among them, the target updated marker points include the target updated marker point j and the target updated marker point k, and the parameter types of the target updated marker point j and the target updated marker point k are the same. The parameter type of the updated marker point is used to indicate which key part of the target object in the updated multimedia data the updated marker point corresponds to.

[0129] Specifically, for example, if the target parameter type is used to indicate the shoulder of the target object, then the computer device obtains the parameter type of each of at least two updated marker points, determines whether each updated marker point is the updated marker point corresponding to the shoulder of the target object, and thus determines the updated marker points corresponding to the shoulder of the target object as the target updated marker points.

[0130] 4. The computer device obtains the third updated pixel position of the target update marker point j in the updated multimedia data and the fourth updated pixel position of the target update marker point k in the updated multimedia data, and determines the target update distance between the target update marker point j and the target update marker point k according to the third updated pixel position and the fourth updated pixel position.

[0131] Specifically, the computer device can establish a rectangular coordinate system with any position in the updated multimedia data as the coordinate origin. It can be known that the way of establishing the coordinate system in the updated multimedia data should be the same as the way of establishing the coordinate system in the multimedia data in the above steps, that is, both establish a rectangular coordinate system with the upper left corner position of the multimedia data as the coordinate origin, or both establish a rectangular coordinate system with any other position of the multimedia data as the coordinate origin. After establishing the coordinate system, the computer device can obtain the position coordinates of the target update marker point j in this coordinate system to obtain the third updated pixel position, and, obtain the position coordinates of the target update marker point k in this coordinate system to obtain the fourth updated pixel position. Further, the computer device can obtain the distance value between the position coordinates corresponding to the third updated pixel position and the position coordinates corresponding to the fourth updated pixel position. For example, it can be the distance value between the abscissas of the two coordinates, to obtain the target update distance between the target update marker point j and the target update marker point k.

[0132] 5. The computer device determines the distance difference between the target initial distance and the target update distance as the marker difference value.

[0133] Optionally, the computer device can obtain the transformation distance between the transformation marker point j and the transformation marker point k corresponding to the shoulder of the target object in the target multimedia data by performing transformation processing on the multimedia data when the expected transformation value corresponding to the transformation trigger instruction is the largest; determine the marker difference value between the initial marker point and the updated marker point according to the target initial distance, the target update distance, and the transformation distance. The specific method for calculating the marker difference value can refer to formula (1-1).

[0134] S205. Determine the detection result of the transformation function according to the marker difference value and the expected transformation value.

[0135] In the embodiments of the present application, the computer device determines the detection result of the transformation function by obtaining the marker difference value between the initial marker point and the updated marker point, and according to the marker difference value and the expected transformation value, so as to determine whether the transformation function is an effective transformation function or an ineffective transformation function. The specific method for determining the detection result of the transformation function according to the marker difference value and the expected transformation value can refer to Figure 3 the description of step S105 in the corresponding embodiment, which will not be elaborated here.

[0136] S206, if the detection result of the transformation function is an invalid result of the transformation function, a function repair request for the transformation function is generated.

[0137] In the embodiment of the present application, the computer device determines whether the absolute difference between the marked difference value and the expected transformation value belongs to the effective error range. If it is determined that the absolute difference between the marked difference value and the expected transformation value does not belong to the effective error range, the computer device determines that the detection result of the transformation function is an invalid result of the transformation function. Among them, the function repair request includes difference data.

[0138] S207, send the function repair request to the user terminal so that the user terminal repairs the transformation function based on the function repair request.

[0139] In the embodiment of the present application, after generating the function repair request, the computer device sends the function repair request to the user terminal so that the user terminal repairs the transformation function based on the function repair request. Since the function repair request includes difference data, the user can view the difference data through the user terminal, thereby determining the size between the difference data and the effective error range according to the difference data, and then realizing the repair of the transformation function so that the repaired transformation function can transform the multimedia data.

[0140] In the embodiment of the present application, when it is determined that the transformation function is an invalid transformation function, by sending the invalid information to the user terminal, the developer of the transformation function can repair the transformation function in time so that the repaired transformation function can beautify the target object, thereby improving the user experience of the personnel using the transformation function.

[0141] The method of the embodiment of the present application is introduced above. Next, the device of the embodiment of the present application is introduced.

[0142] See Figure 7 , Figure 7 is a schematic structural diagram of a composition of a data verification device provided by an embodiment of the present application. The above-mentioned data verification device may be a computer program (including program code) running in a computer device. For example, the data verification device is an application software; the device can be used to execute the corresponding steps in the method provided by the embodiment of the present application. The device 70 includes:

[0143] The first recognition module 71 is configured to, in response to a transformation trigger instruction for a transformation function, recognize an initial marked point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction;

[0144] An object transformation module 72, configured to perform transformation processing on a target object in the multimedia data based on the transformation trigger instruction, to obtain updated multimedia data; the transformation trigger instruction includes an expected transformation value for the transformation processing;

[0145] A second recognition module 73, configured to recognize updated marker points corresponding to the initial marker points in the updated multimedia data;

[0146] A difference acquisition module 74, configured to acquire a marker difference value between the initial marker points and the updated marker points;

[0147] A result acquisition module 75, configured to determine a detection result of the transformation function according to the marker difference value and the expected transformation value.

[0148] Optionally, the first recognition module 71 includes:

[0149] A first object determination unit 711, configured to recognize the multimedia data targeted by the transformation trigger instruction based on a target detection model, to determine a target object in the multimedia data;

[0150] A feature determination unit 712, configured to acquire an object feature of the target object in the multimedia data, and determine an initial marker point corresponding to the target object from the marker points corresponding to the object feature.

[0151] Optionally, the first recognition module 71 includes:

[0152] A parameter acquisition unit 713, configured to acquire target marker parameters associated with the transformation function corresponding to the transformation trigger instruction, and acquire a target marker parameter range corresponding to the target marker parameters;

[0153] An initial data determination unit 714, configured to, in the multimedia data targeted by the transformation trigger instruction, determine pixel points whose pixel parameters belong to the target marker parameter range as initial marker points corresponding to the target object in the multimedia data; initial marker points with the same target marker parameters are initial marker point pairs.

[0154] Optionally, the apparatus 70 further includes:

[0155] A mapping module 76, configured to acquire pixel values of at least two pixel points in the multimedia data, and map the at least two pixel points to a color map based on the pixel values of the at least two pixel points;

[0156] A clustering module 77, configured to perform clustering processing on the at least two pixel points based on positions of the at least two pixel points in the color map, to obtain a multimedia color cluster;

[0157] A range determination module 78 is configured to determine a marking parameter range according to the color atlas values in the color atlas that have a color distance greater than a color difference threshold from the colors of the multimedia color clusters. The marking parameter range includes a target marking parameter range.

[0158] Optionally, the number of the initial marking points is at least two, and the number of the updated marking points is at least two; the difference acquisition module 74 includes:

[0159] An initial data grouping unit 741 is configured to obtain the initial parameters respectively corresponding to each of at least two initial marking points, obtain the parameter types of the initial parameters respectively corresponding to each of the at least two initial marking points, and perform grouping processing on the at least two initial marking points based on the parameter types of the initial parameters to obtain m pairs of initial marking points; the m pairs of initial marking points include an initial marking point pair i, and the initial marking point pair i includes a first initial marking point and a second initial marking point, where m is a positive integer and i is a positive integer;

[0160] An updated data grouping unit 742 is configured to obtain the updated parameters respectively corresponding to each of at least two updated marking points, obtain the parameter types of the updated parameters respectively corresponding to each of the at least two updated marking points, and perform grouping processing on the at least two updated marking points based on the parameter types of the updated parameters to obtain m pairs of updated marking points; the m pairs of updated marking points include an updated marking point pair i, and the updated marking point pair i includes a first updated marking point and a second updated marking point;

[0161] An initial distance determination unit 743 is configured to obtain a first initial pixel position of the first initial marking point in the multimedia data and a second initial pixel position of the second initial marking point in the multimedia data for each initial marking point pair, and determine an initial distance of the initial marking point pair i according to the first initial pixel position and the second initial pixel position;

[0162] An updated distance determination unit 744 is configured to obtain a first updated pixel position of the first updated marking point in the updated multimedia data and a second updated pixel position of the second updated marking point in the updated multimedia data for each updated marking point pair, and determine an updated distance of the updated marking point pair i according to the first updated pixel position and the second updated pixel position;

[0163] A first difference determination unit 745 is configured to determine a distance difference between the initial distance of the initial marking point pair i and the updated distance of the updated marking point pair i as the i-th marking difference value.

[0164] Optionally, the number of the initial marking points is at least two, and the number of the updated marking points is at least two; the difference acquisition module 74 includes:

[0165] An initial data matching unit 746 is configured to obtain a target parameter type corresponding to the transformation function, obtain a parameter type of each of at least two initial marker points, and determine an initial marker point that matches the target parameter type among the at least two initial marker points as a target initial marker point; the target initial marker points include a target initial marker point j and a target initial marker point k, and the parameter types of the target initial marker point j and the target initial marker point k are the same;

[0166] An initial position determination unit 747 is configured to obtain a third initial pixel position of the target initial marker point j in the multimedia data and a fourth initial pixel position of the target initial marker point k in the multimedia data, and determine a target initial distance between the target initial marker point j and the target initial marker point k according to the third initial pixel position and the fourth initial pixel position;

[0167] An updated data matching unit 748 is configured to obtain a parameter type of each of at least two updated marker points, and determine an updated marker point that matches the target parameter type among the at least two updated marker points as a target updated marker point; the target updated marker points include a target updated marker point j and a target updated marker point k, and the parameter types of the target updated marker point j and the target updated marker point k are the same;

[0168] An updated position determination unit 749 is configured to obtain a third updated pixel position of the target updated marker point j in the updated multimedia data and a fourth updated pixel position of the target updated marker point k in the updated multimedia data, and determine a target updated distance between the target updated marker point j and the target updated marker point k according to the third updated pixel position and the fourth updated pixel position;

[0169] A second difference determination unit 7410 is configured to determine a distance difference between the target initial distance and the target updated distance as the marker difference value.

[0170] Optionally, the difference acquisition module 74 includes:

[0171] A difference data acquisition unit 7411 is configured to obtain difference data between the marker difference value and the expected transformation value;

[0172] A first result determination unit 7412 is configured to determine that the detection result of the transformation function is a valid transformation function result if the difference data belongs to a valid error range;

[0173] A second result determination unit 7413 is configured to determine that the detection result of the transformation function is an invalid transformation function result if the difference data does not belong to the valid error range.

[0174] Optionally, the apparatus 70 further includes:

[0175] A request generation module 79, configured to generate a function repair request for the transformation function if the detection result of the transformation function is an invalid transformation function result; the function repair request includes the difference data.

[0176] A request sending module 710, configured to send the function repair request to a user terminal, so that the user terminal repairs the transformation function based on the function repair request.

[0177] It should be noted that Figure 7 For the content not mentioned in the corresponding embodiment, reference can be made to the description of the method embodiment, which will not be elaborated here.

[0178] In the embodiment of the present application, in response to a transformation trigger instruction for a transformation function, an initial marker point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction is identified; the target object in the multimedia data is subjected to transformation processing based on the transformation trigger instruction to obtain updated multimedia data; wherein, the transformation trigger instruction includes an expected transformation value; an updated marker point corresponding to the initial marker point in the updated multimedia data is identified; a marker difference value between the initial marker point and the updated marker point is obtained, and based on the marker difference value and the expected transformation value, the detection result of the transformation function is determined. Since the expected transformation value corresponding to the transformation trigger instruction can represent the theoretical transformation data volume by which the transformation trigger instruction for the transformation function can transform the multimedia data, and the marker difference value can represent the actual changed data volume of the multimedia data after the transformation processing, by comparing the theoretical transformation data volume and the actual changed data volume of the multimedia data, the detection result of the transformation function, that is, determining whether the transformation function is effective or invalid, can be determined; since the marker points in the multimedia data before and after the transformation are identified, and the marker difference value between the initial marker point and the updated marker point, as well as the difference between the marker difference value and the expected transformation value, are compared, the accuracy of data verification can be improved.

[0179] See Figure 8 , Figure 8 is a schematic structural diagram of a computer device provided by an embodiment of the present application. As Figure 8As shown in the figure, the above computer device 80 may include: a processor 801, a network interface 804, and a memory 805. In addition, the above computer device 80 may further include: a user interface 803 and at least one communication bus 802. Among them, the communication bus 802 is used to realize the connection and communication between these components. Among them, the user interface 803 may include a display screen (Display) and a keyboard (Keyboard). Optionally, the user interface 803 may further include a standard wired interface and a wireless interface. The network interface 804 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 805 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory 805 may further be at least one storage device located far from the aforementioned processor 801. As Figure 8 shown, in the memory 805 as a computer-readable storage medium, there may be included an operating system, a network communication module, a user interface module, and a device control application program.

[0180] In Figure 8 the computer device 80 shown in the figure, the network interface 804 can provide network communication functions; while the user interface 803 is mainly used to provide an input interface for users; and the processor 801 can be used to call the device control application program stored in the memory 805 to implement:

[0181] In response to a transformation trigger instruction for a transformation function, identify the initial marking point corresponding to the target object in the multimedia data targeted by the transformation trigger instruction;

[0182] Based on the transformation trigger instruction, perform transformation processing on the target object in the multimedia data to obtain updated multimedia data; the transformation trigger instruction includes a transformation value for the transformation processing;

[0183] Identify the updated marking point corresponding to the initial marking point in the updated multimedia data;

[0184] Obtain the marking difference value between the initial marking point and the updated marking point;

[0185] According to the marking difference value and the expected transformation value, determine the detection result of the transformation function.

[0186] Optionally, when the processor 801 executes to identify the initial marking point corresponding to the target object in the multimedia data targeted by the transformation trigger instruction, it includes:

[0187] Based on a target detection model, identify the multimedia data targeted by the transformation trigger instruction to determine the target object in the multimedia data;

[0188] Obtain the object features of the target object in the multimedia data, and determine the initial marker points corresponding to the target object from the marker points corresponding to the object features.

[0189] Optionally, the processor 801 executes to identify the initial marker points corresponding to the target object in the multimedia data targeted by the transformation trigger instruction, including:

[0190] Obtain the target marker parameters associated with the transformation function corresponding to the transformation trigger instruction, and obtain the target marker parameter range corresponding to the target marker parameters;

[0191] In the multimedia data targeted by the transformation trigger instruction, determine the pixel points whose pixel parameters belong to the target marker parameter range as the initial marker points corresponding to the target object in the multimedia data; the initial marker points with the same target marker parameters are initial marker point pairs.

[0192] Optionally, the processor 801 may call the program code to perform the following operations:

[0193] Obtain the pixel values of at least two pixel points in the multimedia data, and based on the pixel values of the at least two pixel points, map the at least two pixel points to a color map;

[0194] Perform clustering processing on the at least two pixel points based on their positions in the color map to obtain a multimedia color cluster;

[0195] Determine the marker parameter range according to the color map values whose color distances from the multimedia color cluster are greater than the color difference threshold, and the marker parameter range includes the target marker parameter range.

[0196] Optionally, the number of the initial marker points is at least two, and the number of the updated marker points is at least two; the processor 801 executes to obtain the marker difference value between the initial marker points and the updated marker points, including:

[0197] Obtain the initial parameters respectively corresponding to each of the at least two initial marker points, obtain the parameter types of the initial parameters respectively corresponding to each of the at least two initial marker points, and perform grouping processing on the at least two initial marker points based on the parameter types of the initial parameters to obtain m initial marker point pairs; the m initial marker point pairs include the initial marker point pair i, and the initial marker point pair i includes a first initial marker point and a second initial marker point, m is a positive integer, and i is a positive integer;

[0198] Obtain the update parameters corresponding to each of at least two update marker points respectively, obtain the parameter types of the update parameters corresponding to each of these update marker points respectively, and perform grouping processing on the at least two update marker points based on the parameter types of the update parameters to obtain m update marker point pairs; the m update marker point pairs include the update marker point pair i, and the update marker point pair i includes a first update marker point and a second update marker point;

[0199] Obtain the first initial pixel position of the first initial marker point in the multimedia data and the second initial pixel position of the second initial marker point in the multimedia data in each initial marker point pair, and determine the initial distance of the initial marker point pair i according to the first initial pixel position and the second initial pixel position;

[0200] Obtain the first updated pixel position of the first updated marker point in the updated multimedia data and the second updated pixel position of the second updated marker point in the updated multimedia data in each updated marker point pair, and determine the updated distance of the updated marker point pair i according to the first updated pixel position and the second updated pixel position;

[0201] Determine the distance difference between the initial distance of the initial marker point pair i and the updated distance of the updated marker point pair i as the i-th marker difference value.

[0202] Optionally, the number of the initial marker points is at least two, and the number of the updated marker points is at least two; when the processor 801 executes obtaining the marker difference value between the initial marker point and the updated marker point, it includes:

[0203] Obtain the target parameter type corresponding to the transformation function, obtain the parameter types of each of at least two initial marker points, and determine the initial marker points that match the target parameter type among the at least two initial marker points as target initial marker points; the target initial marker points include a target initial marker point j and a target initial marker point k, the parameter types of the target initial marker point j and the target initial marker point k are the same, j is a positive integer, and k is a positive integer;

[0204] Obtain the third initial pixel position of the target initial marker point j in the multimedia data and the fourth initial pixel position of the target initial marker point k in the multimedia data, and determine the target initial distance between the target initial marker point j and the target initial marker point k according to the third initial pixel position and the fourth initial pixel position;

[0205] Obtain the parameter types of each of at least two updated marker points, and determine the updated marker points that match the target parameter type among the at least two updated marker points as target updated marker points; the target updated marker points include target updated marker point j and target updated marker point k, and the parameter types of the target updated marker point j and the target updated marker point k are the same;

[0206] Obtain the third updated pixel position of the target updated marker point j in the updated multimedia data and the fourth updated pixel position of the target updated marker point k in the updated multimedia data, and determine the target updated distance between the target updated marker point j and the target updated marker point k according to the third updated pixel position and the fourth updated pixel position;

[0207] Determine the distance difference between the target initial distance and the target updated distance as the marker difference value.

[0208] Optionally, the processor 801 executes identifying the initial marker points corresponding to the target objects in the multimedia data, including:

[0209] Optionally, the processor 801 executes determining the detection result of the transformation function according to the marker difference value and the expected transformation value, including:

[0210] Obtain the difference data between the marker difference value and the expected transformation value;

[0211] If the difference data belongs to the valid error range, determine the detection result of the transformation function as a valid transformation function result;

[0212] If the difference data does not belong to the valid error range, determine the detection result of the transformation function as an invalid transformation function result.

[0213] Optionally, the processor 801 may call the program code to perform the following operations:

[0214] If the detection result of the transformation function is an invalid transformation function result, generate a function repair request for the transformation function; the function repair request includes the difference data;

[0215] Send the function repair request to the user terminal so that the user terminal repairs the transformation function based on the function repair request.

[0216] It should be understood that the computer device 80 described in the embodiments of the present application may execute the descriptions of the above-mentioned one data verification method in the corresponding embodiments described above Figure 3 and Figure 6 and may also execute the above Figure 7The description of the above-mentioned data verification device in the corresponding embodiment will not be repeated here. In addition, the description of the beneficial effects of using the same method will not be repeated either.

[0217] In an embodiment of the present application, in response to a transformation trigger instruction for a transformation function, an initial marker point corresponding to a target object in the multimedia data targeted by the transformation trigger instruction is identified; the target object in the multimedia data is transformed based on the transformation trigger instruction to obtain updated multimedia data; wherein, the transformation trigger instruction includes an expected transformation value; an updated marker point corresponding to the initial marker point in the updated multimedia data is identified; a marker difference value between the initial marker point and the updated marker point is obtained, and based on the marker difference value and the expected transformation value, a detection result of the transformation function is determined. Since the expected transformation value corresponding to the transformation trigger instruction can represent the theoretical transformation data volume by which the transformation trigger instruction for the transformation function can transform the multimedia data, and the marker difference value can represent the actual changed data volume of the multimedia data after the transformation process, by comparing the theoretical transformation data volume and the actual changed data volume of the multimedia data, the detection result of the transformation function can be determined, that is, it can be determined whether the transformation function is effective or invalid; since the marker points in the multimedia data before and after the transformation are identified, and the marker difference value between the initial marker point and the updated marker point, as well as the difference between the marker difference value and the expected transformation value, are compared, the accuracy of data verification can be improved.

[0218] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the method as described in the foregoing embodiments. The computer may be a part of the above-mentioned computer device. For example, it may be the above-mentioned processor 801. As an example, the program instructions may be deployed to be executed on one computer device, or be deployed to be executed on multiple computer devices located at one place. Or, they may be executed on multiple computer devices distributed at multiple places and interconnected through a communication network. The multiple computer devices distributed at multiple places and interconnected through a communication network may form a blockchain network.

[0219] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it may include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0220] The above-disclosed is only the preferred embodiment of the present application. Of course, it cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A data verification method, characterized in that, Including: Obtain the pixel values of at least two pixel points in the multimedia data, and map the at least two pixel points to a color map based on the pixel values of the at least two pixel points; Perform clustering processing on the at least two pixel points based on the positions of the at least two pixel points in the color map to obtain a multimedia color cluster; Determine a marker parameter range according to the map color values in the color map whose color distances from the multimedia color cluster are greater than a color difference threshold; In response to a transformation trigger instruction for a transformation function, obtain the target marker parameter associated with the transformation function corresponding to the transformation trigger instruction, and obtain the target marker parameter range corresponding to the target marker parameter; the marker parameter range includes the target marker parameter range; In the multimedia data targeted by the transformation trigger instruction, determine the pixel points whose pixel parameters belong to the target marker parameter range as the initial marker points corresponding to the target object in the multimedia data; The initial marker points with the same target marker parameter are initial marker point pairs; Perform transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data; the transformation trigger instruction includes the expected transformation value of the transformation processing; Identify the updated marker points corresponding to the initial marker points in the updated multimedia data; Obtain the marker difference value between the initial marker point and the updated marker point; Determine the detection result of the transformation function according to the marker difference value and the expected transformation value.

2. The method according to claim 1, wherein The number of the initial marker points is at least two, and the number of the updated marker points is at least two; The obtaining the marker difference value between the initial marker point and the updated marker point includes: Obtain the initial parameters respectively corresponding to each of at least two initial marker points, obtain the parameter types of the initial parameters respectively corresponding to each of the at least two initial marker points, and perform grouping processing on the at least two initial marker points based on the parameter types of the initial parameters to obtain m initial marker point pairs; the m initial marker point pairs include the initial marker point pair i, the initial marker point pair i includes a first initial marker point and a second initial marker point, m is a positive integer, and i is a positive integer; Obtain the updated parameters respectively corresponding to each of at least two updated marker points, obtain the parameter types of the updated parameters respectively corresponding to each of the at least two updated marker points, and perform grouping processing on the at least two updated marker points based on the parameter types of the updated parameters to obtain m updated marker point pairs; the m updated marker point pairs include the updated marker point pair i, the updated marker point pair i includes a first updated marker point and a second updated marker point; Obtain the first initial pixel position of the first initial marker point in the multimedia data and the second initial pixel position of the second initial marker point in the multimedia data in each initial marker point pair, and determine the initial distance of the initial marker point pair i according to the first initial pixel position and the second initial pixel position; Obtain the first updated pixel position of the first updated marker point in the updated multimedia data for each pair of updated marker points, and the second updated pixel position of the second updated marker point in the updated multimedia data. Determine the updated distance of the pair of updated marker points i according to the first updated pixel position and the second updated pixel position. Determine the distance difference between the initial distance of the initial marker point pair i and the updated distance of the updated marker point pair i as the i-th marker difference value.

3. The method according to claim 1, wherein The number of the initial marker points is at least two, and the number of the updated marker points is at least two. The obtaining the marker difference value between the initial marker point and the updated marker point includes: Obtain the target parameter type corresponding to the transformation function, obtain the parameter type of each of at least two initial marker points, and determine the initial marker point that matches the target parameter type among the at least two initial marker points as the target initial marker point; the target initial marker points include the target initial marker point j and the target initial marker point k, the parameter types of the target initial marker point j and the target initial marker point k are the same, j is a positive integer, and k is a positive integer. Obtain the third initial pixel position of the target initial marker point j in the multimedia data and the fourth initial pixel position of the target initial marker point k in the multimedia data, and determine the target initial distance between the target initial marker point j and the target initial marker point k according to the third initial pixel position and the fourth initial pixel position. Obtain the parameter type of each of at least two updated marker points, and determine the updated marker point that matches the target parameter type among the at least two updated marker points as the target updated marker point; the target updated marker points include the target updated marker point j and the target updated marker point k, and the parameter types of the target updated marker point j and the target updated marker point k are the same. Obtain the third updated pixel position of the target updated marker point j in the updated multimedia data and the fourth updated pixel position of the target updated marker point k in the updated multimedia data, and determine the target updated distance between the target updated marker point j and the target updated marker point k according to the third updated pixel position and the fourth updated pixel position. Determine the distance difference between the target initial distance and the target updated distance as the marker difference value.

4. The method according to claim 1, wherein The determining the detection result of the transformation function according to the marker difference value and the expected transformation value includes: Obtain the difference data between the marker difference value and the expected transformation value. If the difference data belongs to the effective error range, determine that the detection result of the transformation function is a valid result of the transformation function. If the difference data does not belong to the effective error range, determine that the detection result of the transformation function is an invalid result of the transformation function.

5. The method according to claim 4, wherein The method further includes: If the detection result of the transformation function is an invalid result of the transformation function, generate a function repair request for the transformation function; the function repair request includes the difference data. Send the function repair request to the user terminal so that the user terminal repairs the transformation function based on the function repair request.

6. A data verification device, characterized in that, The device includes: A mapping module, configured to obtain pixel values of at least two pixel points in the multimedia data, and map the at least two pixel points to a color map based on the pixel values of the at least two pixel points; A clustering module, configured to perform clustering processing on the at least two pixel points based on the positions of the at least two pixel points in the color map to obtain a multimedia color cluster; A range determination module, configured to determine a marked parameter range according to the map color values in the color map whose color distances from the multimedia color cluster are greater than a color difference threshold; A first recognition module, configured to, in response to a transformation trigger instruction for a transformation function, obtain a target marked parameter associated with the transformation function corresponding to the transformation trigger instruction, and obtain a target marked parameter range corresponding to the target marked parameter; the marked parameter range includes the target marked parameter range; The first recognition module is further configured to, in the multimedia data targeted by the transformation trigger instruction, determine the pixel points whose pixel parameters belong to the target marked parameter range as the initial marked points corresponding to the target object in the multimedia data; the initial marked points with the same target marked parameter are initial marked point pairs; An object transformation module, configured to perform transformation processing on the target object in the multimedia data based on the transformation trigger instruction to obtain updated multimedia data; the transformation trigger instruction includes an expected transformation value of the transformation processing; A second recognition module, configured to recognize updated marked points corresponding to the initial marked points in the updated multimedia data; A difference acquisition module, configured to acquire a marked difference value between the initial marked points and the updated marked points; A result acquisition module, configured to determine a detection result of the transformation function according to the marked difference value and the expected transformation value.

7. A computer device, characterized in that, Includes: A processor, a memory, and a network interface; The processor is connected to the memory and the network interface, wherein the network interface is used to provide a data communication function, the memory is used to store program codes, and the processor is used to call the program codes to execute the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, and the computer program is suitable for being loaded and executed by the processor to execute the method according to any one of claims 1-5.

Images