Medium image generation method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a medium image generation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a medium image design instruction; querying target medium elements of at least two element types matched with the medium image design instruction in a medium element library, and obtaining a target typesetting scheme; the target typesetting scheme is a medium element typesetting scheme determined by analyzing a typesetting mode of medium elements in the reference medium image; the reference medium image is a medium image to be referenced indicated by the medium image design instruction; and combining the target medium elements according to the target typesetting scheme to generate a target medium image meeting the medium image design instruction. By adopting the method, the medium image can be generated more intelligently.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a media image generation method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] In today's digital age, artificial intelligence generated content (AIGC) technology is gradually becoming a new engine for the creative industry. AIGC technology uses artificial intelligence technologies such as machine learning, computer vision and natural language processing to automatically generate content such as text, images, music and videos.

[0003] In the creative fields of digital marketing and media data (such as advertising), the diversity and innovation of media image content are crucial to attracting the audience's attention. More and more companies and creators are trying to use AI tools to generate media image materials. However, AIGC technology in related technologies can usually only generate a single media element and cannot stably produce a diverse set of media images at the same time. When dealing with complex projects, a lot of manual intervention is still required, resulting in low creative efficiency and easy design duplication.

[0004] Therefore, the media image generation method in the related art has the problem of low intelligence. Summary of the invention

[0005] Based on this, it is necessary to provide a media image generation method, apparatus, computer device, computer-readable storage medium and computer program product that can generate media images more intelligently in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for generating a media image, comprising:

[0007] Get media image design instructions;

[0008] Querying the media element library for at least two target media elements of element types that match the media image design instruction, and obtaining a target layout scheme; the target layout scheme is a media element layout scheme determined by parsing the layout method of the media elements in the reference media image; the reference media image is the media image that the media image design instruction indicates needs to be referenced;

[0009] The target media elements are combined according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

[0010] In one embodiment, when the media image design instruction carries a reference image, obtaining the target layout scheme includes:

[0011] Acquire a reference image carried in the media image design instruction, and use the reference image as the reference media image;

[0012] The layout of the media elements in the reference media image is parsed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

[0013] In one embodiment, when the media image design instruction carries image content description text, obtaining the target layout scheme includes:

[0014] Searching for a target historical media image that matches the image content description text in the historical media image library as the reference media image; the delivery effect corresponding to the target historical media image meets a preset condition;

[0015] The layout of the media elements in the reference media image is parsed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

[0016] In one embodiment, there are multiple target media images, and the method further includes:

[0017] In response to a parameter adjustment operation on any target media element in any target media image, obtaining a media element configuration parameter input by the parameter adjustment operation;

[0018] According to the media element configuration parameters, similar media elements in all the target media images are adjusted in batches to obtain new target media images; the similar media elements and any of the target media elements belong to the same element type.

[0019] In one embodiment, the method further comprises:

[0020] Acquire historical media elements in historical media images, and expand the content of the historical media elements through a pre-trained content generation model to generate expanded media elements; wherein the pre-trained content generation model controls the base template generation style of each of the expanded media elements through preset parameters and templates;

[0021] The historical media element and the expanded media element are added to the media element library.

[0022] In one embodiment, searching the media element library for target media elements of at least two element types matching the media image design instruction includes:

[0023] Parsing the media image design instruction by natural language processing technology, searching the media element library for at least two types of media elements that match the media image design intent of the media image design instruction as candidate media elements;

[0024] In response to a screening operation on the candidate media elements, the candidate media elements screened out by the screening operation are used as the target media elements.

[0025] In a second aspect, the present application further provides a media image generating device, comprising:

[0026] An instruction acquisition module, used for acquiring media image design instructions;

[0027] A scheme acquisition module is used to query the media element library for at least two target media elements of element types that match the media image design instruction, and to acquire a target typesetting scheme; the target typesetting scheme is a media element typesetting scheme determined by parsing the typesetting method of the media elements in the reference media image; the reference media image is a media image that the media image design instruction indicates needs to be referenced;

[0028] The image generation module is used to combine the target media elements according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

[0029] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0030] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0031] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0032] The above-mentioned media image generation method, device, computer equipment, computer-readable storage medium and computer program product obtain media image design instructions; query at least two element types of target media elements that match the media image design instructions in the media element library, and obtain a target typesetting scheme; the target typesetting scheme is a media element typesetting scheme determined by parsing the typesetting method of the media elements in the reference media image; the reference media image is the media image that the media image design instructions indicate needs to be referenced; and the target media elements are combined according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

[0033] In this way, by querying the media element library for at least two element types that match the media image design instructions, and obtaining the target layout scheme, the target layout scheme is a media element layout scheme determined by parsing the layout method of the media elements of the reference media image that the media image design instructions indicate needs to refer to, so as to realize intelligent recommendation of media elements and layout schemes that meet the user's design needs, and then combine the target media elements according to the target layout scheme, so as to realize intelligent completion of the filling, layout and overall style optimization of the media elements, generate a target media image that meets the media image design instructions, and ensure that the generated target media image complies with the user's original media image design instructions, that is, ensure that the generated target media image meets the user's design needs, thereby realizing full process automation and intelligence from design to output, thereby improving the intelligence of the media image generation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 is a schematic flow chart of a method for generating a media image in one embodiment;

[0036] Figure 2 An interactive interface diagram of a media image generation method in one embodiment;

[0037] Figure 3 A schematic diagram of a process of generating media elements of different element types in one embodiment;

[0038] Figure 4 It is an interactive interface diagram of a media image generation method in another embodiment;

[0039] Figure 5This is an interactive interface diagram of yet another method for generating a media image in another embodiment;

[0040] Figure 6 An interactive interface diagram of another media image generation method in one embodiment;

[0041] Figure 7 An interactive interface diagram of another media image generation method in another embodiment;

[0042] Figure 8 is a flow chart of yet another method for generating a media image in one embodiment;

[0043] Fig. 9 is a schematic flow chart of a method for generating a media image in another embodiment;

[0044] Fig.10 is a structural block diagram of a media image generating device in one embodiment;

[0045] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0048] In one embodiment, Figure 1 As shown, a method for generating a media image is provided. This embodiment uses the method applied to a computer device as an example. It can be understood that the computer device can be a terminal, a server, or a system including a terminal and a server. In this embodiment, the method includes the following steps:

[0049] Step S110, obtaining a media image design instruction.

[0050] The media image design instruction may be an instruction to instruct the media image design system to perform media image design.

[0051] In practical applications, media image design instructions may include specific requirements and descriptions for guiding media image design related operations, which can be used to characterize media image design requirements (media image design intent).

[0052] Among them, media images can refer to advertising images in actual applications.

[0053] In a specific implementation, a user account may input media image design instructions in an instruction input page provided by a media image design system for media image design.

[0054] Step S120, searching the media element library for target media elements of at least two element types that match the media image design instruction, and obtaining a target layout solution.

[0055] The target layout scheme is a media element layout scheme determined by analyzing the layout method of media elements in the reference media image.

[0056] Among them, the typesetting scheme of the media elements can define the layout arrangement of various media elements in the picture for generating the media image.

[0057] The reference media image is a media image that the media image design instruction indicates needs to be referenced.

[0058] Among them, the media element library includes media elements generated by a pre-trained content generation model.

[0059] Among them, the pre-trained content generation model can be a generative artificial intelligence (AIGC) model.

[0060] The media elements may refer to the materials (materials) that make up the media image. For example, the media elements may include different element types such as background, model, copy, button, decorative sticker, LOGO, etc.

[0061] In a specific implementation, a computer device can respond to an input media image design instruction, query a media element library for at least two element types of target media elements that match the media image design instruction, and obtain a target layout scheme that matches the media image design instruction. The target layout scheme is a media element layout scheme determined by the computer device by parsing the layout method of the media elements in a reference media image. The reference media image is a media image that the media image design instruction indicates needs to be referenced.

[0062] Step S130 , combining target media elements according to the target typesetting scheme to generate a target media image that meets the media image design requirements represented by the media image design instructions.

[0063] In a specific implementation, a computer device can combine target media elements according to a target layout scheme. Specifically, it can complete element filling, layout, and overall style optimization according to the target layout scheme to generate a target media image that meets the media image design requirements represented by the media image design instructions.

[0064] Furthermore, this embodiment can generate multiple target media images in batches, and by generating a large number of multi-version target media images with the same frame, the matching effect of different media elements can be tested. Specifically, the computer device can parse the media image design instructions through natural language processing technology, determine the number of image designs, combine the target media elements according to the target layout plan, and generate multiple target media images that meet the number of image designs.

[0065] Among them, multiple target layout schemes matching the media image design instructions can be obtained, and each target media element is combined through each target layout scheme to generate multiple target media images that meet the image design quantity.

[0066] In some embodiments, the target media elements of the same element type may include multiple ones, and the target media elements of different element types may be combined through a target layout scheme to generate multiple target media images that meet the number of image designs.

[0067] In this way, the target media elements under different element types are combined through the target layout scheme, and the filling, layout and overall style optimization of the elements are intelligently completed. Multiple versions of target media images with the same frame can be generated in batches, ensuring the consistency of design quality and style. There is no reliance on manual replacement of media elements one by one to synthesize different media images, which reduces the user's manual operations in the design process and improves the efficiency and convenience of mass production.

[0068] In the above-mentioned media image generation method, media image design instructions are obtained; target media elements of at least two element types that match the media image design instructions are queried in the media element library, and a target layout scheme is obtained; the target layout scheme is a media element layout scheme determined by parsing the layout method of media elements in a reference media image; the reference media image is a media image that the media image design instructions indicate needs to be referenced; the target media elements are combined according to the target layout scheme to generate a target media image that meets the media image design instructions.

[0069] In this way, by querying the media element library for at least two element types that match the media image design instructions, and obtaining the target layout scheme, the target layout scheme is a media element layout scheme determined by parsing the layout method of the media elements of the reference media image that the media image design instructions indicate needs to refer to, so as to realize intelligent recommendation of media elements and layout schemes that meet the user's design needs, and then combine the target media elements according to the target layout scheme, so as to realize intelligent completion of the filling, layout and overall style optimization of the media elements, generate a target media image that meets the media image design instructions, and ensure that the generated target media image complies with the user's original media image design instructions, that is, ensure that the generated target media image meets the user's design needs, thereby realizing full process automation and intelligence from design to output, thereby improving the intelligence of the media image generation method.

[0070] In one embodiment, when a media image design instruction carries a reference image, a target layout scheme matching the media image design instruction is obtained, including: obtaining the reference image carried by the media image design instruction, and using the reference image as a reference media image; parsing the layout method of media elements in the reference media image, and determining the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

[0071] In some scenarios, the media image design instruction may carry a reference image. For example, the media image design system provides an "upload image to make the same style" function. When a user uses the "upload image to make the same style" function to input a media image design instruction, the media image design instruction carries a reference image uploaded by the user account.

[0072] Among them, the design layout parameters can be used to define the layout, arrangement and visual relationship of each target media element in the picture.

[0073] For example, design layout parameters may include element position and distribution parameters (coordinate positioning, alignment, distribution rules), visual hierarchy parameters (element size level, color contrast level, overlap and occlusion level), element spacing and white space parameters, and other parameters.

[0074] Among them, the element size level can reflect the visual primary and secondary and hierarchical relationship by setting the size of different elements. Among them, the color contrast level refers to the use of color brightness (brightness), purity (saturation), and hue differences to create visual hierarchy. Among them, the overlap and occlusion level is used to specify whether there is overlap between elements and the order of overlap to reflect the sense of space and depth.

[0075] In this way, when the computer device is in the process of obtaining the target layout scheme that matches the media image design instructions, the computer device can obtain the reference image carried by the media image design instructions, and use the reference image as a reference media image, and determine the design layout parameters indicated by the media image design instructions by parsing the layout method of the media elements in the reference media image; the design layout parameters can be used to characterize the target layout scheme, so as to combine the target media elements according to the target layout scheme, thereby realizing the element layout format of the reference image uploaded by the user account, and producing the target media image that meets the media image design requirements represented by the media image design instructions.

[0076] The technical solution of this embodiment is, when the media image design instruction carries a reference image, by obtaining the reference image carried by the media image design instruction and using the reference image as the reference media image; parsing the typesetting method of the media elements in the reference media image, and determining the design typesetting parameters indicated by the media image design instruction; the design typesetting parameters are used to characterize the target typesetting scheme. In this way, by parsing the typesetting method of the media elements of the reference image carried by the media image design instruction, the design typesetting parameters indicated by the media image design instruction are determined, so that the selected target media elements can be combined through the design typesetting parameters, and the element typesetting format of the reference image uploaded by the user account can be replicated to produce the target media image that meets the media image design instruction, so that the target media image better meets the user's image design needs.

[0077] In one embodiment, when a media image design instruction carries an image content description text, a target layout scheme is obtained, including: searching a historical media image library for a target historical media image that matches the image content description text as a reference media image; the delivery effect corresponding to the target historical media image meets a preset condition; the delivery effect corresponding to the target historical media image meets a preset condition; the layout method of the media elements in the reference media image is parsed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

[0078] The image content description text may refer to text used to describe the image content of the media image to be generated, such as "20 online dating pictures describing in-person relationships are needed" or "design a retro-style sports shoe advertising picture".

[0079] Among them, the historical media image library can be a database storing historical media images.

[0080] Among them, historical media images may include media images that have been historically produced by the media image design system, and may also include media images that have been uploaded by various user accounts.

[0081] In a specific implementation, when a user account inputs a media image design instruction into a media image design system, the user account may also input an image content description text, that is, the media image design instruction carries the image content description text, and the computer device may respond to the media image design instruction and query a target historical media image that matches the image content description text in the historical media image library as a reference media image. The delivery effect corresponding to the target historical media image satisfies a preset condition.

[0082] Among them, the computer device can first search the historical media image library for historical media images that match the image content description text as candidate historical media images, and then compare the delivery effects corresponding to each candidate historical media image, and select the candidate historical media image whose delivery effect meets the preset conditions as the target historical media image.

[0083] For example, if the image content description text indicates that the current advertisement image to be designed is a food-related advertisement image, the computer device will query the historical media image library for historical media images related to food as candidate historical media images, and then compare the delivery effects of each candidate historical media image to filter out the target historical media image.

[0084] The delivery effect can be measured by at least one delivery index such as click-through rate, conversion rate, exposure and interaction, so as to select candidate historical media images that meet a preset threshold value as target historical media images according to the index data of each candidate historical media image under each delivery index. The number of target historical media images can be at least one.

[0085] In practical applications, the screened target historical media images can also be named historical running volume advertising images.

[0086] In this way, the target historical media image is used as a reference media image, the layout method of the media elements in the reference media image is analyzed, and the design layout parameters indicated by the media image design instructions are determined; the design layout parameters are used to characterize the target layout scheme.

[0087] The technical solution of this embodiment is that, when the media image design instruction carries the image content description text, the target historical media image that matches the image content description text is searched in the historical media image library as the reference media image; the delivery effect corresponding to the target historical media image meets the preset conditions; the typesetting method of the media elements in the reference media image is parsed to determine the design typesetting parameters indicated by the media image design instruction; the design typesetting parameters are used to characterize the target typesetting scheme. In this way, by referring to the typesetting method of the target historical media image whose delivery effect meets the preset conditions and whose corresponding image content conforms to the image content description text, the design typesetting parameters that characterize the target typesetting scheme are determined, so as to achieve the use of effective design elements, layout logic, etc. in the target historical media image with better delivery effect, reduce the risk of design errors, improve the quality of the designed media image, and thus improve the delivery effect of the generated media image and the utilization efficiency of advertising resources.

[0088] In one embodiment, there are multiple target media images, and the method further includes: in response to a parameter adjustment operation on any target media element in any target media image, obtaining a media element configuration parameter input for the parameter adjustment operation; and batch adjusting similar media elements in all target media images according to the media element configuration parameters to obtain new target media images.

[0089] The same type of media element and any target media element belong to the same element type.

[0090] The parameter adjustment operation may refer to an operation for adjusting the parameters of a media element, for example, adjusting the current position, size, angle, color, etc. of a media element, or selecting a new media element to replace the current media element.

[0091] The media element configuration parameters only refer to the new configuration parameters input by the parameter adjustment operation for the media element to be adjusted. For example, if the parameter adjustment operation is to adjust the position of a certain media element, the media element configuration parameters input by the parameter adjustment operation may include new position parameters (for example, which may be represented by X and Y coordinates); for example, if the parameter adjustment operation is to adjust the size of a certain media element, the media element configuration parameters input by the parameter adjustment operation may include the new size.

[0092] In a specific implementation, in order to further improve the efficiency of media image design, the present application can not only batch generate multiple media images at one time in the production process, but also provide a batch editing function, allowing the user account to quickly replace and synthesize similar media elements in each target media image generated. Specifically, the user account can adjust the parameters of any target media element in any target media image among the generated target media images, and by inputting the parameter adjustment operation of any target media element in any target media image into the computer device, the computer device can respond to the parameter adjustment operation of any target media element in any target media image and obtain the media element configuration parameters input by the parameter adjustment operation; in this way, the computer device can batch adjust the similar media elements in all target media images according to the input media element configuration parameters to obtain new target media images, wherein the similar media elements belong to the same element type as the any target media element.

[0093] For example, if the user account wants to adjust the "text media element", such as adjusting the position, size, angle, color and other styles of the text media element, such as Figure 2 As shown, an interactive interface diagram of a media image generation method is provided, in which a user account can input a parameter adjustment operation for a copy media element in any target media image. After the computer device receives the parameter adjustment operation for the copy media element in any target media image by the user account, the copy media elements in all target media images can be adjusted in batches according to the media element configuration parameters input by the parameter adjustment operation. In this way, the user account can efficiently and uniformly edit and adjust a series of target media images without having to process them one by one, thereby improving the efficiency of batch editing.

[0094] The technical solution of this embodiment is that there are multiple target media images. By responding to the parameter adjustment operation of any target media element in any target media image, the media element configuration parameters input by the parameter adjustment operation are obtained; according to the media element configuration parameters, the same type of media elements in all target media images are batch adjusted to obtain new target media images; the same type of media elements and any target media element belong to the same element type. In this way, the user account can quickly adjust the media elements and layout of a series of target media images, so that the user account can efficiently and uniformly edit and adjust a series of target media images without having to process them one by one, thereby improving the efficiency of batch editing.

[0095] In one embodiment, the method also includes: obtaining historical media elements in historical media images, expanding the content of the historical media elements through a pre-trained content generation model, and generating extended media elements; wherein the pre-trained content generation model controls the base template generation style of each extended media element through preset parameters and templates; and adding the historical media elements and the extended media elements to a media element library.

[0096] Among them, the pre-trained content generation model can be a generative artificial intelligence (AIGC) model.

[0097] Among them, historical media elements may refer to media elements that constitute historical media images.

[0098] In a specific implementation, a computer device can obtain historical media elements in historical media images, expand the content of the historical media elements through a pre-trained content generation model to generate expanded media elements, and add the historical media elements and the expanded media elements to a media element library to enrich the media element content.

[0099] In the process of expanding the content of the historical media elements by using the pre-trained content generation model to generate the expanded media elements, the content of the historical media elements belonging to the first element type can be expanded by using the pre-trained content generation model. For example, the first element type can include element types such as background, model, and copy.

[0100] The technical solution of this embodiment obtains historical media elements in historical media images, expands the content of the historical media elements through a pre-trained content generation model, and generates expanded media elements; wherein the pre-trained content generation model controls the base mold generation style of each expanded media element through preset parameters and templates; and adds the historical media elements and the expanded media elements to the media element library. In this way, by expanding the content of historical media elements through a pre-trained content generation model, the content of the media elements can be enriched and the diversity of the media elements in the media element library can be improved. At the same time, by controlling the base mold generation style of each expanded media element through preset parameters and templates through the pre-trained content generation model, the quality of the expanded media elements can be guaranteed.

[0101] In order to facilitate the understanding of those skilled in the art, Figure 3 A schematic diagram of the process of generating media elements of different element types is provided. Figure 3As shown, taking the advertising image as an example, the media elements in the advertising image include 6 types of elements, including background, model, copy, logo, button, and decorative map. All of them are generated by a comprehensive AIGC model and cannot meet the complex requirements. Therefore, the best acquisition method can be used for each element type. For example, the AIGC model is used to generate the media elements of the first element type, such as background, model, and copy, and decorative maps (the second element type) are supplemented by citing network materials, and the logo (the third element type) is filled in by manual uploading to obtain various media elements. At the same time, the user account can select the media elements generated by the AIGC model, and select the media elements that meet the requirements and add them to the media element library.

[0102] In this way, the most suitable generation or acquisition method can be used for different types of media elements (such as backgrounds, models, copywriting, etc.), rather than using large prompts with poor controllability to generate a complete media image at one time, thereby ensuring the quality and diversity of each media element.

[0103] Among them, when using AIGC technology (such as ComfyUI, Stable Diffusion, LLM (Large Language Mode)) to automatically generate backgrounds, models and texts in advertising images, the base template generation style can be controlled through preset parameters and templates. After generation, the AI ​​scoring model will evaluate the feasibility of the generated media elements to ensure that the media elements generated by any prompt meet the design requirements. At the same time, it can also take into account personalized demands, allowing user accounts to modify and supplement prompts according to personalized needs to generate media elements that meet design requirements.

[0104] In order to facilitate the understanding of those skilled in the art, Figure 4 An interactive interface diagram of a media image generation method is provided, which can generate model media elements using AIGC technology. Figure 4 As shown, the user account can enter a prompt word to instruct ComfyUI to generate a model element that meets the requirements.

[0105] In this way, AIGC technology is used to train the base mold of each media element to ensure its lower quality limit, and it supports modifying the prompt to generate personalized media elements. The AI ​​scoring model is combined to evaluate the generation results to improve the upper quality limit, protect all AIGC-generated media elements to comply with design principles, and improve the quality of the generated media elements.

[0106] Among them, you can use ComfyUI and the flower font copywriting system to reserve commonly used flower font copywriting styles. The LLM large model automatically generates flower font copywriting and applies preset generated copywriting (picture format, and because the generation process is completed by the system in a streamlined manner, the copywriting can be modified at any time).

[0107] Among them, ComfyUI and Stable Diffusion technologies can be used to generate background images and models through text-generated images and image-generated images. Cutting out is a separate execution workflow, and the generated model media elements can be cut out.

[0108] Among them, the fancy font generation is created by the template background system. The template preset template is created and the fancy font content is generated. The stroke, blur and other styles can be adjusted. In order to facilitate the understanding of those skilled in the art, Figure 5 An interactive interface diagram of a media image generation method is provided. Figure 5 As shown, the user account can adjust the flower font preset template in the interactive interface diagram.

[0109] Among them, all media images produced in this system have production process information and media delivery data, and retain the original media elements and layout information of all media images, which can be managed and analyzed in the media element library. At the same time, all media elements can be recommended to the production process for efficient reuse. For example, the system can extract the logo required for media image design instructions and other media elements (backgrounds, models, text, buttons, decorative maps) that have been used in the past from the media element library to realize AI recommendation reuse to enrich the visual effects of the generated media images. In order to facilitate the understanding of technicians in this field, Figure 6 An interactive interface diagram of a media image generation method is provided, such as Figure 6 As shown, media elements can be recommended in the interactive interface diagram.

[0110] In addition, this model can be customized according to different design themes and styles with a high degree of customization. This automated process reduces the steps of manual participation and improves production efficiency.

[0111] In practice, the functionality of existing AIGC tools can be expanded through custom development to better adapt them to specific design needs and workflows. For example, this can include integrating third-party services or developing specific plug-ins to enhance element management and automated workflow capabilities.

[0112] In one embodiment, target media elements of at least two element types that match the media image design instructions are queried in the media element library, including: parsing the media image design instructions through natural language processing technology, and querying the media element library for at least two element types of media elements that match the media image design intent of the media image design instructions as candidate media elements; in response to a screening operation on the candidate media elements, the candidate media elements screened out by the screening operation are used as target media elements.

[0113] The screening operation on the candidate media elements may refer to an operation in which the user screens out the desired media elements from the candidate media elements.

[0114] In a specific implementation, when a computer device queries a media element library for target media elements of at least two element types that match a media image design instruction, it can query the media element library for media elements of at least two element types that match the media image design intent of the media image design instruction as candidate media elements.

[0115] In some embodiments, when the media image design instructions carry image content description text, the image content description text can be parsed by natural language processing technology, and at least two types of media elements that match the media image design intent can be screened out according to the parsing results as candidate media elements. If the media image design instructions instruct to select media images with good advertising data performance as the same style, the system can intelligently match the media elements in the library that meet the requirements, such as color, shape, style, etc., and automatically replace the media elements for synthesis to achieve rapid reproduction. If the design requirements specify that a specific font or pattern is required, the system can still continue to understand the intent and modify it, identify the semantics and select the most appropriate media element from the media element library for replacement.

[0116] Furthermore, for example, the user account inputs "design a retro-style sports shoe advertising picture, with sepia as the main color tone and an old-fashioned texture as the overall picture". This instruction clearly specifies that the style of the media image to be designed is retro style, and gives specific requirements for style-related media elements such as color tone and picture texture. After receiving such a media image design instruction, the system will match various media elements with sepia tones and old-fashioned textures in the media element library.

[0117] For another example, if a user inputs "design a retro-style perfume advertising picture", the system will understand the key semantics of "retro style" and "perfume", and then filter out relevant media elements with retro features in the media element library, such as old-fashioned perfume bottle shapes, retro patterns, and old-fashioned paper backgrounds for recommendation.

[0118] In other embodiments, when a user account uploads a reference image, for example, if the user account uses the "upload image to make the same style" function, the computer device can use computer vision technology to identify and analyze the media elements in the reference image carried by the media image design instructions, and intelligently recommend media elements that meet the media image design intent (such as color, shape, style, etc. that meet the media image design intent) from the media element library as candidate media elements.

[0119] For example, computer vision technology can be used to perform color analysis on reference images and extract the main color distribution in the reference images, so as to select media elements with similar color distribution to the reference images in the media element library as candidate media elements. For example, pixels are classified according to color similarity through clustering algorithms to determine the main colors (such as blue and yellow used in large areas in the reference images) and auxiliary colors. At the same time, the position and function of these colors in the reference images are analyzed, such as whether certain colors are used to highlight the subject, create an atmosphere, or serve as a background. Then, media elements with matching or coordinated colors are selected from the media element library as candidate media elements, such as shapes, patterns, and other media elements whose background colors are close to the main colors of the reference images, to ensure the consistency of the overall color style.

[0120] You can also use color theory knowledge to determine the color matching rules followed by the reference image, such as complementary color matching (such as red and green, blue and yellow), adjacent color matching (such as matching adjacent color ranges of red, orange and yellow) or three-color matching, etc. According to the corresponding color matching rules, search for matching media elements in the media element library as candidate media elements. If the reference image uses blue and yellow complementary colors to create a strong visual impact, then when recommending media elements, you will also give priority to media elements that can be integrated into this complementary color system, such as blue shape media elements with yellow decorative lines.

[0121] The shape detection algorithm of computer vision can also be used to identify the main shape categories (such as circles, squares, triangles, etc.) that appear in the reference image, as well as their combination forms, proportional relationships and other shape features. For example, the main part of the reference image is a shape of multiple nested circles, or a layout with a square as a frame. Then search for media elements with similar shape features in the media element library as candidate media elements. If it is found that there are shape media elements in the media element library that are also nested in circles or can be adapted to the square frame, they will be recommended as candidate media elements so that the shape style characteristics of the reference image can be continued in the new design.

[0122] You can also consider the spatial layout of the shapes in the reference image, such as whether the shapes are symmetrically distributed, staggered, or radially arranged. Select shape media elements from the media element library based on this layout feature so that they can be combined in a similar layout in the new design, thereby matching the overall shape layout style shown in the reference image and creating a similar visual effect.

[0123] The style characteristics of the reference images can also be quantitatively analyzed from multiple dimensions, such as the degree of realism of the picture (hyperrealism, realism or abstract style), the characteristics of the lines (simple and tough straight line style, or soft and curved line style), texture performance (smooth texture, rough texture or matte texture, etc.) and light and shadow effects (strong light contrast, soft light and shadow or backlight effect, etc.).

[0124] Based on these quantified style features, media elements with similar style features are matched in the media element library as candidate media elements. For example, if the reference image is in a minimalist style with soft light and shadow and simple lines, then those patterns and decorative media elements in the media element library with simple lines, soft light and shadow effects and an overall minimalist style will be screened out as candidate media elements.

[0125] You can also pre-label various media elements in the media element library with style labels, such as retro style, modern minimalist style, cartoon style, etc., and use computer vision technology to determine the style label category of the reference image, such as determining that the reference image belongs to the retro style. The system will then select media elements that are also marked as retro style from the media element library as candidate media elements for recommendation, and further accurately match specific media elements that better meet the requirements based on the more specific performance of the reference image in the retro style (such as European retro, Chinese retro, etc.), such as media elements with European retro patterns.

[0126] In some embodiments, the media image design instruction may carry a reference image and image content description text, and the computer device may identify and analyze the media elements in the reference image through computer vision technology, and parse the image content description text through natural language processing technology, thereby searching the media element library for at least two element types of media elements that match the media image design intent of the media image design instruction as candidate media elements. Figure 7 As shown, an interactive interface diagram of a media image generation method is provided, in which a user account can upload a reference image and select an image style and text information. At the same time, a text describing the image content can be input in the interactive interface diagram. In addition, the similarity between the media image designed and generated by the system and the reference image can be further selected, and the number of media images required to be designed and generated can be indicated.

[0127] For example, if a user account uploads a reference image with unique layout and color matching, and the instruction requires the addition of sports elements, the system will analyze the visual features of the reference image (such as color distribution, layout, etc.), and combine it with the understanding of "sports elements" to recommend media elements with coordinated colors, reasonable layout and in line with the sports theme as candidate media elements, such as sports equipment, images of athletes, etc., so that the recommended candidate media elements can not only continue the style of the reference image, but also meet the new theme requirements.

[0128] For another example, the user account inputs "Design an advertising picture for a seaside tourist attraction in City A. Refer to the style of this reference image and add some water sports elements." The reference image is a landscape photo of City B, and the overall tone of the picture is blue and white. The system can analyze the color distribution of the reference image, and then, combined with the image content description text, recommend similar visual media elements in the seaside attractions in City A, such as scene pictures with azure blue waters and white buildings of a unified style on the seaside as the main background media elements. For "water sports", media elements such as sailboats and jet skis are recommended, and the colors and styles of these media elements will be adapted with reference to the reference image.

[0129] It is understandable that when there are no specific design requirements in the image content description text, such as "need 20 online dating pictures describing meeting in person", "design a retro-style sports shoe advertisement picture". In some embodiments, the computer device may also search for target historical media images matching the image content description text in the historical media image library, analyze the media elements in the target historical media images through computer vision technology, and select media elements similar to the media elements in the target historical media images in color, style, shape, etc. in the media element library as candidate media elements.

[0130] In this way, candidate media elements can be displayed in the system for selection by the user account. The user account inputs a screening operation on the candidate media elements to the computer device according to actual needs. The computer device can use the candidate media elements screened out by the screening operation as the target media elements.

[0131] The technical solution of this embodiment uses natural language processing technology to parse media image design instructions, and searches for at least two types of media elements that match the media image design intent of the media image design instructions in the media element library as candidate media elements; in response to a screening operation on the candidate media elements, the candidate media elements screened out by the screening operation are used as target media elements. In this way, by parsing media image design instructions through natural language processing technology, complex media image design instructions can be understood, and media elements that meet the design intent can be intelligently recommended from the media element library for selection by the user account, thereby obtaining target media elements that meet the user's design needs, effectively improving the intelligence and flexibility of media element recommendation.

[0132] It is understandable that, in other embodiments, the media image design instructions carry image content description text, and the media image design instructions can also be parsed through natural language processing technology and converted into specific design layout parameters.

[0133] For example, a user account inputs "Make a cosmetics advertising picture, place the product picture in the top center, arrange the product name and efficacy introduction text in two lines below the product picture, the text color is pink, and the overall text layout should be centered." This instruction focuses on the layout of each media element in the picture and the color of the text and other details. The system will parse the media image design instructions based on natural language processing technology and convert them into specific design layout parameters.

[0134] In this way, natural language processing technology can be used to accurately understand the design intent of the media image, determine the design layout parameters that meet the design intent of the media image, and improve the accuracy of the design layout parameters.

[0135] In other embodiments, in the process of obtaining the target layout scheme, multiple historical running volume advertising images whose delivery effects meet preset conditions can be pre-selected. Each historical running volume advertising image can be a variety of scenes, such as sports scenes, travel scenes, family scenes, etc. According to the layout method of the media elements in each historical running volume advertising image, a candidate media element layout scheme is obtained, and each candidate media element layout scheme is displayed on the system, such as Figure 4 As shown, multiple candidate media element layout schemes can be displayed in the interactive interface diagram for selection by the user account. The user account selects the desired candidate media element layout scheme as the target layout scheme. Specifically, the computer device can respond to the user account's selection operation on the candidate media element layout scheme and use the candidate media element layout scheme selected by the selection operation as the target layout scheme.

[0136] In this way, by providing media element layout plans corresponding to multiple historical advertising images whose delivery effects meet preset conditions, the user account can select the required media element layout plan as the target layout plan. This not only ensures the quality of the target layout plan, but also allows the target layout plan to meet the media image design intentions of the user account, thereby achieving more intelligent acquisition of the target layout plan.

[0137] In another embodiment, Figure 8 As shown, a flow chart of a media image generation method is provided. Taking an advertisement image as an example, the media elements in the advertisement image include 6 types of elements, including background, model, copy, logo, button, and decorative map. All of them are generated by a comprehensive AIGC model, which cannot meet the complex requirements. Therefore, the best acquisition method can be used for each element type. For example, the AIGC model is used to generate the media elements of the first element type, such as background, model, and copy, and the decorative map (the second element type) is supplemented by citing network materials. The logo (the third element type) is filled in by manual uploading to obtain various media elements, and each element is handed over to AI for recommended layout (scaling size, placement, layer relationship, etc.) and complete the final creation.

[0138] In practical applications, you can develop dedicated scripts and plug-ins to automate specific design tasks. By creating automated scripts, you can achieve automatic filling and layout of media elements.

[0139] In this way, this application builds a highly automated workflow for generating media elements, intelligent filling and typesetting, optimizing the overall style, and batch editing and modification, making the media image design process easier. The user account only needs to use the "upload pictures to make the same style" function or use a simple description, such as "need 20 online dating pictures to describe the real-life relationship", and the system can use NLP technology to understand the media image design instructions entered by the user account, automatically identify the needs and trigger the query of at least two element types of target media elements that match the media image design instructions, and obtain the target typesetting scheme, and intelligently complete the filling, typesetting and overall style optimization of the target media elements. Finally, the generated results are manually confirmed and modification suggestions are made. The system performs batch editing and adjustments to ensure that each target media image complies with the user's original instructions, thereby realizing full process automation from design to output, reducing the user's manual operations, and making the entire process from media element generation to typesetting optimization smoother and more efficient.

[0140] In another embodiment, Fig. 9 As shown, a method for generating a media image is provided, which is described by taking the method applied to a computer device as an example, and includes the following steps:

[0141] Step S902, obtaining historical media elements in the historical media image, and expanding the content of the historical media elements through a pre-trained content generation model to generate expanded media elements.

[0142] Step S904: adding the historical media elements and the expanded media elements to the media element library.

[0143] Step S906, obtaining a media image design instruction, parsing the media image design instruction through natural language processing technology, and searching the media element library for at least two element types of media elements that match the media image design intent of the media image design instruction as candidate media elements.

[0144] Step S908, in response to the screening operation on the candidate media elements, the candidate media elements screened out by the screening operation are used as target media elements.

[0145] Step S910, parsing the layout of the media elements in the reference media image to determine the design layout parameters indicated by the media image design instruction.

[0146] Step S912, combining target media elements according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

[0147] Step S914, in response to a parameter adjustment operation on any target media element in any target media image, obtaining media element configuration parameters input by the parameter adjustment operation.

[0148] Step S916: perform batch adjustment on the same type of media elements in all target media images according to the media element configuration parameters to obtain new target media images.

[0149] It should be noted that the specific limitations of the above steps can refer to the specific limitations of a media image generation method mentioned above.

[0150] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0151] Based on the same inventive concept, the embodiment of the present application also provides a media image generation device for implementing the media image generation method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more media image generation device embodiments provided below can refer to the limitations of the media image generation method above, and will not be repeated here.

[0152] In an exemplary embodiment, Fig.10 As shown, a medium image generation device is provided, including: an instruction acquisition module 1010, a solution acquisition module 1020 and an image generation module 1030, wherein:

[0153] The instruction acquisition module 1010 is used to acquire the media image design instruction.

[0154] The scheme acquisition module 1020 is used to query the media element library for at least two element types of target media elements that match the media image design instructions, and to obtain a target layout scheme; the target layout scheme is a media element layout scheme determined by parsing the layout method of the media elements in the reference media image; the reference media image is the media image that the media image design instructions indicate needs to be referenced.

[0155] The image generation module 1030 is used to combine the target media elements according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

[0156] In one of the embodiments, when the media image design instruction carries a reference image, the scheme acquisition module 1020 is specifically used to acquire the reference image carried by the media image design instruction, and use the reference image as the reference media image; parse the layout method of the media elements in the reference media image, and determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

[0157] In one of the embodiments, when the media image design instruction carries an image content description text, the solution acquisition module 1020 is specifically used to query the historical media image library for a target historical media image that matches the image content description text as the reference media image; the delivery effect corresponding to the target historical media image meets a preset condition; the layout method of the media elements in the reference media image is analyzed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout solution.

[0158] In one embodiment, there are multiple target media images, and the device further includes: an adjustment module, used to respond to a parameter adjustment operation on any target media element in any of the target media images, and obtain the media element configuration parameters input by the parameter adjustment operation; based on the media element configuration parameters, batch adjust the same type of media elements in all the target media images to obtain new target media images; the same type of media elements and any of the target media elements belong to the same element type.

[0159] In one embodiment, the device also includes: an element generation module, used to obtain historical media elements in historical media images, expand the content of the historical media elements through a pre-trained content generation model, and generate extended media elements; wherein the pre-trained content generation model controls the base template generation style of each of the extended media elements through preset parameters and templates; and adds the historical media elements and the extended media elements to the media element library.

[0160] In one of the embodiments, the solution acquisition module 1020 is specifically used to parse the media image design instruction through natural language processing technology, query at least two element types of media elements that match the media image design intent of the media image design instruction in the media element library as candidate media elements; in response to a screening operation on the candidate media elements, use the candidate media elements screened out by the screening operation as the target media elements.

[0161] Each module in the above-mentioned media image generation device can be implemented in whole or in part by software, hardware or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0162] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.11As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for generating a media image is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0163] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0164] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0165] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0166] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0167] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0168] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0169] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0170] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for generating a media image, characterized in that: The method comprises: Get media image design instructions; Querying the media element library for at least two target media elements of element types that match the media image design instruction, and obtaining a target layout scheme; the target layout scheme is a media element layout scheme determined by parsing the layout method of the media elements in the reference media image; the reference media image is the media image that the media image design instruction indicates needs to be referenced; The target media elements are combined according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

2. The method according to claim 1, characterized in that In the case where the media image design instruction carries a reference image, the step of obtaining a target layout scheme includes: Acquire a reference image carried in the media image design instruction, and use the reference image as the reference media image; The layout of the media elements in the reference media image is parsed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

3. The method according to claim 1, characterized in that In the case where the media image design instruction carries image content description text, the step of obtaining a target layout scheme includes: Searching for a target historical media image that matches the image content description text in the historical media image library as the reference media image; the delivery effect corresponding to the target historical media image meets a preset condition; The layout of the media elements in the reference media image is parsed to determine the design layout parameters indicated by the media image design instruction; the design layout parameters are used to characterize the target layout scheme.

4. The method according to claim 1, characterized in that There are multiple target media images, and the method further includes: In response to a parameter adjustment operation on any target media element in any target media image, obtaining a media element configuration parameter input by the parameter adjustment operation; According to the media element configuration parameters, similar media elements in all the target media images are adjusted in batches to obtain new target media images; the similar media elements and any of the target media elements belong to the same element type.

5. The method according to claim 1, characterized in that The method further comprises: Acquire historical media elements in historical media images, and expand the content of the historical media elements through a pre-trained content generation model to generate expanded media elements; wherein the pre-trained content generation model controls the base template generation style of each of the expanded media elements through preset parameters and templates; The historical media element and the expanded media element are added to the media element library.

6. The method according to claim 1, characterized in that The step of searching the media element library for target media elements of at least two element types matching the media image design instruction includes: Parsing the media image design instruction by natural language processing technology, searching the media element library for at least two types of media elements that match the media image design intent of the media image design instruction as candidate media elements; In response to a screening operation on the candidate media elements, the candidate media elements screened out by the screening operation are used as the target media elements.

7. A media image generating device, characterized in that: The device comprises: An instruction acquisition module, used for acquiring media image design instructions; A scheme acquisition module is used to query the media element library for at least two target media elements of element types that match the media image design instruction, and to acquire a target typesetting scheme; the target typesetting scheme is a media element typesetting scheme determined by parsing the typesetting method of the media elements in the reference media image; the reference media image is a media image that the media image design instruction indicates needs to be referenced; The image generation module is used to combine the target media elements according to the target typesetting scheme to generate a target media image that meets the media image design instructions.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images