Figure artistic image generation method based on AIGC and voice prompt

Abstract

The invention relates to the technical field of artistic image generation, in particular to a figure artistic image generation method based on AIGC and voice prompt, and the method comprises the steps: obtaining historical work data generated through the historical operation of a user, carrying out the preprocessing of the historical work data, and outputting the feature data of the historical work; according to the historical work feature data, analyzing a first emotion preference value and a second emotion preference value of the user on the to-be-generated image of the current operation to obtain a comprehensive preference value of the user on the to-be-generated image of the current operation; and comparing the comprehensive preference value of the user on the to-be-generated image in the current operation with a preference value threshold, and when the comprehensive preference value is greater than the preference value threshold, outputting the to-be-generated image. Whether the to-be-generated artistic works conform to the preferences of the user or not is judged according to the comprehensive preferences of the user on the artistic works, the operability is good, the accuracy is high, and the calculation amount is small.

Description

Background Art

[0002] AIGC refers to the generation of multimedia content such as text, images, videos, and audio through artificial intelligence technology. In recent years, with the development of deep learning, generative adversarial networks (GANs), and diffusion models, AIGC has been widely used in the creative field. In terms of image generation, AIGC relies on multimodal learning, that is, combining different types of data (such as text, voice, images, etc.) to automatically generate high-quality artistic images through deep learning models. This provides a technical basis for generating artistic images of characters from voice descriptions.

[0003] Problems in the prior art: Currently, there is no method to judge whether the artwork to be generated is in line with the user's preferences based on the user's comprehensive preferences for artworks. Based on this problem, a method for generating a character artistic image based on AIGC and voice prompts is proposed. Summary of the invention

[0004] The main purpose of the present invention is to provide a method for generating an artistic image of a character based on AIGC and voice prompts, so as to effectively solve the above-mentioned problems mentioned in the background technology.

[0005] The technical solution of the present invention is as follows:

[0006] A method for generating artistic image of a person based on AIGC and voice prompts is proposed. The method comprises the following steps:

[0007] S1. Obtain historical work data generated by historical operations of the user, pre-process the historical work data, and output historical work feature data;

[0008] S2. Analyze the first emotion preference value and the second emotion preference value of the user for the image to be generated in this operation according to the historical work feature data, and obtain the comprehensive preference value of the user for the image to be generated in this operation;

[0009] S3. Compare the user's comprehensive preference value for the image to be generated in this operation with the preference value threshold. When the comprehensive preference value is greater than the preference value threshold, output the image to be generated.

[0010] A further improvement of the present invention is that the historical work data in S1 includes historical images generated by historical operations of the user and voice prompts corresponding to the historical images.

[0011] A further improvement of the present invention is that S1 comprises the following specific steps:

[0012] S11, extracting the historical images generated by the user's n historical operations;

[0013] S12, obtaining the average value Ri of all pixel values ​​of a single historical image, the covariance Ci of all pixel values ​​of a single historical image, and the standard deviation Di of all pixel values ​​of a single historical image;

[0014] S13, calculate the mean value of the pixel values ​​of the historical image pixels The covariance mean of all historical image pixel values The mean standard deviation of the pixel values ​​of all historical images

[0015] A further improvement of the present invention is that S1 further includes:

[0016] S14, extract the voice prompts corresponding to the historical images generated by the user's n historical operations, import the voice prompts into the text recognition engine to convert them into text prompts, divide the text prompts according to the sentence positions to obtain multiple entries, calculate the frequency value of a single entry in the text prompt, extract the entries with a frequency value greater than a preset frequency threshold as keywords, form a keyword set, and generate a keyword set sequence corresponding to the voice prompt (P 1 ,P 2 ,…,P n );

[0017] S15. Obtain the number of elements Ni in the keyword set corresponding to a single voice prompt, where Ni represents the number of elements in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation, and calculate the average of the number of elements in n keyword sets.

[0018] S16, identifying the emotional strength S of the elements in the keyword set corresponding to a single voice prompt ij , S ij Indicates the emotional intensity of the jth element in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation in history, and calculates the average emotional intensity of the elements in a single keyword set Among them, m represents the number of element types in a single keyword set, n j represents the number of j-th elements,

[0019] S17. Calculate the average sentiment intensity of elements in the n keyword set

[0020] A further improvement of the present invention is that S2 includes the following specific steps: calculating the first emotion preference value of the user for the image to be generated in this operation, and the calculation formula of the first emotion preference value is: Among them, Rt is the average value of all pixel values ​​of the image to be generated in this operation, Ct is the covariance of the pixel values ​​of all pixel points of the image to be generated in this operation, Dt is the standard deviation of the pixel values ​​of all pixel points of the image to be generated in this operation, a and b are the influencing factors of the user's first emotional preference value for the image to be generated in this operation, and a and b are both greater than 0.

[0021] A further improvement of the present invention is that S2 further includes:

[0022] S22, calculating the second emotion preference value of the user for the image to be generated in this operation, the second emotion preference value calculation formula is: Among them, Nt is the number of keyword set elements in the voice prompt corresponding to the image to be generated in this operation, and St is the emotional strength value of the keyword set elements in the voice prompt corresponding to the image to be generated in this operation;

[0023] S23, calculating the user's comprehensive preference value for the image to be generated in this operation, the comprehensive preference value calculation formula is: in, is the first sentiment preference value ratio coefficient, is the coefficient of the second sentiment preference value.

[0024] The technical effects of the present invention are as follows:

[0025] A method for generating artistic images of characters based on AIGC and voice prompts is constructed. The present invention obtains historical work data generated by historical operations of users, pre-processes the historical work data, and outputs historical work feature data; according to the historical work feature data, the first emotional preference value and the second emotional preference value of the user for the image to be generated in this operation are analyzed to obtain the user's comprehensive preference value for the image to be generated in this operation; the user's comprehensive preference value for the image to be generated in this operation is compared with the preference value threshold, and when the comprehensive preference value is greater than the preference value threshold, the image to be generated is output. The method judges whether the artwork to be generated is in line with the user's preferences based on the user's comprehensive preference for the artwork, and has good operability, high accuracy, and low calculation amount. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0027] Figure 1 The figure is a flow chart of the method for generating an artistic image of a person based on AIGC and voice prompts of the present invention. DETAILED DESCRIPTION

[0028] The present invention aims to propose a method for generating an artistic image of a person based on AIGC and voice prompts. The method obtains historical work data generated by historical operations of the user, pre-processes the historical work data, and outputs historical work feature data. According to the historical work feature data, the first emotion preference value and the second emotion preference value of the user for the image to be generated in this operation are analyzed to obtain the user's comprehensive preference value for the image to be generated in this operation. The user's comprehensive preference value for the image to be generated in this operation is compared with the preference value threshold, and when the comprehensive preference value is greater than the preference value threshold, the image to be generated is output. The method judges whether the artwork to be generated is in line with the user's preference based on the user's comprehensive preference for the artwork, and has good operability, high accuracy, and low calculation amount.

[0029] Embodiment 1:

[0030] This embodiment proposes a method for generating an artistic image of a character based on AIGC and voice prompts, such as Figure 1 As shown, the specific steps include:

[0031] S1. Obtain historical work data generated by historical operations of the user, pre-process the historical work data, and output historical work feature data;

[0032] S2. Analyze the first emotion preference value and the second emotion preference value of the user for the image to be generated in this operation according to the historical work feature data, and obtain the comprehensive preference value of the user for the image to be generated in this operation;

[0033] S3. Compare the user's comprehensive preference value for the image to be generated in this operation with the preference value threshold. When the comprehensive preference value is greater than the preference value threshold, output the image to be generated.

[0034] In this embodiment, the historical work data in S1 includes historical images generated by historical operations of the user and voice prompts corresponding to the historical images.

[0035] In this embodiment, S1 includes the following specific steps:

[0036] S11, extracting historical images generated by n historical operations of the user: The system first extracts n historical images generated by the user from the user's historical operation records. These images are the final results of the user's past operations and are an important basis for understanding the user's preferences.

[0037] S12. Obtain the average value Ri of the pixel values ​​of all pixels of a single historical image, the covariance Ci of the pixel values ​​of all pixels of a single historical image, and the standard deviation Di of the pixel values ​​of all pixels of a single historical image; the average value of the pixel values ​​of all pixels of a single historical image represents the balanced value of the overall brightness or color characteristics of the image; the covariance of the pixel values ​​of all pixels of a single historical image is used to capture the correlation characteristics of the changes in pixel values ​​in the image; the standard deviation of the pixel values ​​of all pixels of a single historical image represents the degree of dispersion of pixel values ​​in the image, reflecting the contrast or color diversity of the image.

[0038] S13, calculate the mean value of the pixel values ​​of the historical image pixels The covariance mean of all historical image pixel values The mean standard deviation of the pixel values ​​of all historical images These means are used to reflect the user's overall visual preference for images and serve as a reference standard for subsequent image generation.

[0039] In this embodiment, the S1 further includes:

[0040] S14, extract the voice prompts corresponding to the historical images generated by the user's n historical operations, import the voice prompts into the text recognition engine to convert them into text prompts, divide the text prompts according to the sentence positions to obtain multiple entries, calculate the frequency value of a single entry in the text prompt, extract the entries with a frequency value greater than a preset frequency threshold as keywords, form a keyword set, and generate a keyword set sequence corresponding to the voice prompt (P 1 ,P 2 ,…,P n );

[0041] S15. Obtain the number of elements Ni in the keyword set corresponding to a single voice prompt, where Ni represents the number of elements in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation, and calculate the average of the number of elements in n keyword sets.

[0042] S16, identifying the emotional strength S of the elements in the keyword set corresponding to a single voice prompt ij , S ij Indicates the emotional intensity of the jth element in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation in history, and calculates the average emotional intensity of the elements in a single keyword set Among them, m represents the number of element types in a single keyword set, n j represents the number of j-th elements,

[0043] S17. Calculate the average sentiment intensity of elements in the n keyword set

[0044] In this embodiment, S2 includes the following specific steps: calculating the user's first emotion preference value for the image to be generated in this operation, and the calculation formula of the first emotion preference value is: Among them, Rt is the average value of all pixel values ​​of the image to be generated in this operation, Ct is the covariance of all pixel values ​​of the image to be generated in this operation, Dt is the standard deviation of all pixel values ​​of the image to be generated in this operation, a and b are the influencing factors of the user's first emotional preference value for the image to be generated in this operation, and a and b are both greater than 0. This formula reflects the matching degree between the pixel features of the image to be generated and the features of the user's historical images, thereby inferring the user's visual preference for the new image.

[0045] In this embodiment, S2 further includes:

[0046] S22, calculating the second emotion preference value of the user for the image to be generated in this operation, the second emotion preference value calculation formula is: Among them, Nt is the number of keyword set elements in the voice prompt corresponding to the image to be generated in this operation, and St is the emotional strength value of the keyword set elements in the voice prompt corresponding to the image to be generated in this operation; this formula reflects the difference between the emotional characteristics of the image to be generated and the emotional preferences of the user in historical operations. If the difference is small, the user's emotional preference value for the image will be higher.

[0047] S23, calculating the user's comprehensive preference value for the image to be generated in this operation, the comprehensive preference value calculation formula is: in, is the first sentiment preference value ratio coefficient, is the second emotion preference value ratio coefficient. According to the specific application scenario, the weights of the two coefficients can be adjusted to reflect the different attention paid by users to visual and emotional preferences.

[0048] Embodiment 2:

[0049] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-mentioned method for generating an artistic image of a character based on AIGC and voice prompts by calling the computer program stored in the memory.

[0050] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the method for generating an artistic image of a character based on AIGC and voice prompts provided in the above method embodiment. The electronic device may also include other components for realizing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface for data input and output. This embodiment will not be described in detail here.

[0051] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), and it can also be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0052] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0053] The present invention is described with reference to flowcharts and block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process or block in the flowchart and block diagram, as well as the combination of processes and blocks in the flowchart or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 Process or multiple processes and boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0054] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 Process or multiple processes and boxes Figure 1 The steps for the functions specified in one or more boxes.

[0055] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for generating an artistic image of a person based on AIGC and voice prompts, characterized in that: The specific steps include: S1. Obtain historical work data generated by historical operations of the user, pre-process the historical work data, and output historical work feature data; S2. Analyze the first emotion preference value and the second emotion preference value of the user for the image to be generated in this operation according to the historical work feature data, and obtain the comprehensive preference value of the user for the image to be generated in this operation; S3. Compare the user's comprehensive preference value for the image to be generated in this operation with the preference value threshold. When the comprehensive preference value is greater than the preference value threshold, output the image to be generated.

2. The method for generating an artistic character image based on AIGC and voice prompt according to claim 1, characterized in that: The historical work data in S1 includes historical images generated by user historical operations and voice prompts corresponding to the historical images.

3. The method for generating an artistic character image based on AIGC and voice prompt according to claim 2, characterized in that: The S1 comprises the following specific steps: S11, extracting the historical images generated by the user's n historical operations; S12, obtaining the average value Ri of all pixel values ​​of a single historical image, the covariance Ci of all pixel values ​​of a single historical image, and the standard deviation Di of all pixel values ​​of a single historical image; S13, calculate the mean value of the pixel values ​​of the historical image pixels The covariance mean of all historical image pixel values The mean standard deviation of the pixel values ​​of all historical images 4. The method for generating an artistic character image based on AIGC and voice prompt according to claim 3, characterized in that: The S1 further comprises: S14, extract the voice prompts corresponding to the historical images generated by the user's n historical operations, import the voice prompts into the text recognition engine to convert them into text prompts, divide the text prompts according to the sentence position to obtain multiple entries, calculate the frequency value of a single entry in the text prompt, extract the entries with a frequency value greater than a preset frequency threshold as keywords, form a keyword set, and generate a keyword set sequence (P1, P2, ..., P n ); S15. Obtain the number of elements Ni in the keyword set corresponding to a single voice prompt, where Ni represents the number of elements in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation, and calculate the average of the number of elements in n keyword sets. S16, identifying the emotional strength S of the elements in the keyword set corresponding to a single voice prompt ij , S ij Indicates the emotional intensity of the jth element in the keyword set corresponding to the voice prompt of the historical image generated by the i-th operation in history, and calculates the average emotional intensity of the elements in a single keyword set Among them, m represents the number of element types in a single keyword set, n j represents the number of j-th elements, S17. Calculate the average sentiment intensity of elements in the n keyword set 5. The method for generating an artistic character image based on AIGC and voice prompt according to claim 4, characterized in that: S2 includes the following specific steps: calculating the user's first emotion preference value for the image to be generated in this operation, and the calculation formula of the first emotion preference value is: Among them, Rt is the average value of all pixel values ​​of the image to be generated in this operation, Ct is the covariance of the pixel values ​​of all pixel points of the image to be generated in this operation, Dt is the standard deviation of the pixel values ​​of all pixel points of the image to be generated in this operation, a and b are the influencing factors of the user's first emotional preference value for the image to be generated in this operation, and a and b are both greater than 0.

6. The method for generating artistic character images based on AIGC and voice prompts according to claim 5, characterized in that: The S2 further includes: S22, calculating the second emotion preference value of the user for the image to be generated in this operation, the second emotion preference value calculation formula is: Among them, Nt is the number of keyword set elements in the voice prompt corresponding to the image to be generated in this operation, and St is the emotional strength value of the keyword set elements in the voice prompt corresponding to the image to be generated in this operation; S23, calculating the user's comprehensive preference value for the image to be generated in this operation, the comprehensive preference value calculation formula is: in, is the first sentiment preference value ratio coefficient, is the coefficient of the second sentiment preference value.

Images