Icon generation method, medium, computer equipment and program product
By inputting image and text description information combined with the target image generation model, the icons with morphological characteristics and target elements are generated, which solves the problem of time-consuming and improved efficiency.
Patent Information
- Application Number
- CN202410825884.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-24
- Publication Date
- 2024-11-12
AI Technical Summary
In the prior art, the icon production process takes a long time and is less efficient, especially the icon design of complex visual features requires a lot of time and effort.
By obtaining input image and text description information of the target icon, a pre-trained target image generation model generates icons with morphological features and target elements, reducing the need for hand-drawing and digital drawing.
It realizes the generation of icons quickly and accurately, reducing the time-consuming process and improving the efficiency of icon production.
Smart Images

Figure CN118939153A_ABST
Abstract
Description
Technical Field
[0001] The present specification relates to the field of image processing technology, and in particular to an icon generation method, medium, computer device and program product. Background Art
[0002] Currently, icons are widely used in all walks of life. In related technologies, high-quality icons are usually produced by professional art designers through hand-drawing, digital drawing, etc. In order to improve the uniqueness and aesthetics of icons, art designers often add complex visual features to icons. However, the more complex the visual features of an icon, the more time and effort it takes to make the icon, resulting in a longer and less efficient icon production process. Summary of the invention
[0003] In a first aspect, an embodiment of the present specification provides an icon generation method, the method comprising: obtaining an input image including a target icon, the input image being used to describe the morphological features of the target icon; obtaining text description information corresponding to the target icon, the text description information being used to describe a target element on the target icon; processing the input image and the text description information by a pre-trained target image generation model to obtain an output image, the output image including the target icon having the morphological features and including the target element.
[0004] In a second aspect, an embodiment of this specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of this specification.
[0005] In a third aspect, an embodiment of this specification provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of this specification when executing the program.
[0006] In a fourth aspect, an embodiment of this specification provides a computer program product, including a computer program, which implements the method described in any embodiment of this specification when executed by a processor.
[0007] In an embodiment of the present specification, the morphological features of the target icon are constrained by inputting an image, and the target elements on the target icon are constrained by the text description information corresponding to the target icon, and then the target image and the text description information are input into the target image generation model, so that the target image generation model generates a target icon having the above-mentioned morphological features and including the target elements. On the one hand, the morphology of the target icon is often complex or irregular, and the morphological features of the target icon can be described more intuitively and accurately by the input image, which facilitates the target image generation model to accurately determine the morphology of the target icon in the output image; on the other hand, since the target element is directly described by the text description information, it does not need to be obtained by drawing, and the process of obtaining the text description information is much more efficient than the process of directly drawing the target element, therefore, various target elements can be quickly generated for the target icon. In summary, the scheme of the embodiment of the present specification can quickly and accurately generate the target icon, reduce the time consumption of the icon production process, and improve the efficiency of the icon production process.
[0008] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the specification and are used together with the specification to illustrate the technical solutions of the specification.
[0010] Figure 1 It is a flowchart of the icon generation method of the embodiment of this specification.
[0011] Figure 2 Schematic diagram of an input image according to an embodiment of the present specification.
[0012] Figure 3 It is a schematic diagram of a user interface of an embodiment of this specification.
[0013] Figure 4 Schematic diagram of an output image according to an embodiment of the present specification.
[0014] Figure 5 is a schematic diagram of a computer device according to an embodiment of the present specification. DETAILED DESCRIPTION
[0015] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this specification. Instead, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.
[0016] The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit this specification. The singular forms of "a", "said" and "the" used in this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items. In addition, the term "at least one" herein means any combination of at least two of any one or more of a plurality of.
[0017] It should be understood that although the terms first, second, third, etc. may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".
[0018] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of this specification and to make the above-mentioned purposes, features and advantages of the embodiments of this specification more obvious and understandable, the technical solutions in the embodiments of this specification are further described in detail below in conjunction with the accompanying drawings.
[0019] At present, icons are widely used in all walks of life. In the field of games, icons are used to represent different game functions, items or control elements of the game interface. For example, props, skills, task signs, menu buttons, etc. in the game are visually represented by icons, which help players quickly identify and operate. In mobile application development, icons are used to identify the application itself and its various functional modules. For example, application icons on the home screen, function buttons in the navigation bar, notification bar icons, etc., these icons not only play a functional role, but also help brand recognition. In web design and software interface design, icons are widely used to represent menu items, action buttons, links, file types, etc. They can make the interface more intuitive and easy to navigate, while improving the user's operating efficiency and experience quality. In educational applications and learning platforms, icons are used to represent different subjects, course content, tools and resources. For example, in educational applications, each subject or course may have a dedicated icon to help students quickly identify and navigate to the required learning content. Icons can also be used in other fields, which will not be listed here one by one.
[0020] In the related art, high-quality icons are usually produced by professional art designers through hand-drawing, digital drawing, etc. For example, icons include some elements such as light and shadow effects, textures, color gradient effects, or line and shape conversions, which may need to be hand-drawn to ensure visual accuracy and beauty. However, the above icon production process is time-consuming and inefficient.
[0021] Based on this, the embodiment of this specification proposes an icon generation method, which constrains the morphological features of the target icon by inputting an image, and constrains the target elements on the target icon by the text description information corresponding to the target icon, and then the target image generation model generates a target icon having the above morphological features and including the target elements based on the target image and the text description information. The above method does not require complex hand-drawing and digital drawing of the target icon, thereby improving the efficiency of the icon production process. The specific implementation method of the embodiment of this specification is illustrated below.
[0022] See also Figure 1 , the icon generation method of the embodiment of this specification includes:
[0023] Step S12: acquiring an input image including a target icon, where the input image is used to describe the morphological features of the target icon;
[0024] Step S14: obtaining text description information corresponding to the target icon, where the text description information is used to describe the target element on the target icon;
[0025] Step S16: Process the input image and text description information through a pre-trained target image generation model to obtain an output image, wherein the output image includes a target icon having the above-mentioned morphological features and including the above-mentioned target elements.
[0026] In step S12, an input image may be acquired. The input image may be an image acquired in real time, or an image acquired and stored in advance. The input image may be an image actually acquired by an image acquisition device, a synthesized digital image, a hand-drawn image, or other types of images. The input image includes a target icon, which may be various types of icons. For example, in the field of games, the target icon may include but is not limited to a prop icon, a skill icon, a task icon, and / or a menu icon in a game; in the field of mobile application development, the target icon may include but is not limited to an application icon on the main screen of an electronic device, a function icon in a navigation bar, and / or a notification bar icon, etc. In other examples, the target icon may also be other types of icons, which are not listed here one by one.
[0027] The input image is used to describe the morphological features of the target icon. The morphological features include but are not limited to at least one of the following: shape features, structural features, contour features, size and proportion features, etc. Among them, the shape features are used to represent the overall shape of the target icon, such as square, circle, triangle or a combination of multiple shapes, or an irregular shape. The structural features are used to represent the visual components and construction methods of the target icon. When the target icon includes multiple target elements, the structural features can represent the spatial layout and arrangement between the target elements. Contour features are used to represent the contour line features of the target icon when viewed from different angles. Size and proportion features are used to represent the size ratio relationship of the various parts of the target icon, such as length, width, height, etc. Figure 2 As shown, the input image depicts the contour features (such as the curvature and length of the contour line of the gem icon) and shape features (such as the area occupied by the icon, the aspect ratio, etc.) of a gem icon.
[0028] In practical applications, the morphology of the target icon is often complex. For example, the shape of the target icon is often irregular, or the outline of the target icon may be integrated into a streamlined design or an irregular edge of a hand-painted style, or the structure of the target icon incorporates a multi-level and dimensional design concept, not limited to the performance of a two-dimensional plane. In this case, the use of an input image can clearly, intuitively and accurately describe the morphological characteristics of the target icon. In addition, since the input image only needs to describe the morphological characteristics of the target icon, and does not need to describe the detailed information of the target icon, the difficulty of obtaining the input image is reduced. For example, a user can draw an icon including a specific shape by hand-drawing, which can effectively reduce the time-consuming process of icon drawing compared to the traditional icon production method in which the user draws the complete shape and specific details of the icon. Of course, the hand-drawing method is only an example, and the input image can also be obtained by other methods. It can be understood that no matter how the input image is obtained, since only the morphological characteristics of the target icon need to be described in the input image, no other information of the target icon is required, therefore, the scheme of the embodiment of this specification has low requirements for the input image, thereby reducing the difficulty of obtaining the input image.
[0029] In some embodiments, the input image is also used to describe the color characteristics of the target icon. The target icon in the input image can be colored by hand-drawing or image rendering, so that the input image can describe the color characteristics of the target icon. In this embodiment, the input image is used to describe the color characteristics of the target icon, which can more accurately represent the color characteristics of the target icon and facilitate the expression of the color change details of the target icon. Especially when the color design of the target icon is more complex (for example, including color gradients or irregular color changes), the color characteristics of the target icon can be intuitively and accurately represented.
[0030] In some embodiments, the target icon includes multiple tiles. The morphological features of different tiles can be the same or different. Figure 2 , the gem icon in the input image includes multiple color blocks of different sizes and shapes. The user can draw multiple blocks included in the target icon in the input image by hand-drawing or digital drawing. By drawing multiple blocks, on the one hand, the input image can more specifically show the morphological characteristics of the target icon; on the other hand, the process of drawing blocks is relatively simple, which will not significantly reduce the efficiency of the icon production process.
[0031] In the example where the target icon includes multiple tiles, the input image is specifically used to describe the color characteristics of each of the multiple tiles. For example, each tile can be colored by hand-drawing or image rendering, so that the input image can describe the color characteristics of each tile. Among them, the color characteristics of each tile can be the same or different. By coloring each tile, the input image can describe the color characteristics of the target icon more finely, thereby improving the visual effect of the generated target icon.
[0032] In step S14, the text description information corresponding to the target icon can be obtained. The text description information can be manually input by the user, or randomly selected from a pre-established text data block corresponding to the type of the target icon, or can be obtained by calling a preset interface (API). The text description information can also be obtained by other methods, which are not listed here. In the embodiment where the user manually inputs the text description information, a user interface 10 can be provided, such as Figure 3 As shown, the user interface 10 may include an input box 20 , in which the user may input text description information.
[0033] The text description information is used to describe the target elements on the target icon. The target elements include, but are not limited to, light and shadow effects, textures, color gradient effects, and / or lines and shapes. Light and shadow effects can add a sense of three-dimensionality and depth to the icon, making it look more vivid. For example, the top of the target icon may have a slight highlight and a shadow on the edge. This light and shadow effect can simulate the effect of sunlight shining from above, so that even on a two-dimensional plane, it can give people a three-dimensional visual experience. Texture refers to the surface texture of the icon, which can increase the sense of reality or a specific style. For example, an embroidery texture can be added to the target icon to give the target icon an embroidery texture. The color gradient effect can make the icon color transition naturally and increase the visual appeal. In the example where the input image is also used to describe the color characteristics of the target icon, the color gradient effect can be added or adjusted for the target icon in the input image through the text description information. Lines and shapes can make the target icon have richer detail information and become more vivid. For example, in the case where the target icon is a wood icon, the shape of flowers or grass can be added to the wood to make the target icon look more lively.
[0034] The text description information may be for the target icon as a whole, or for a part of the target icon, or both. The text description information for the target icon as a whole can be used to describe the overall features of the target icon, such as category, color, shape, etc., for example, “a screenshot of an email client interface including an email envelope icon, showing icons of an inbox and an unread email”, “the target icon is a green icon”, “the target icon is in the shape of a five-pointed star”. The description information for a part of the target icon can be used to describe the local features of the target icon, for example, “the lower left corner of the icon includes a flower”, “the color of the upper right corner of the image gradually changes from green to red”, etc.
[0035] In an embodiment where the input image includes multiple tiles, the text description information may include text description information corresponding to at least one of the multiple tiles. The text description information corresponding to different tiles may be the same or different. For example, assuming that the target element described by the text description information includes the light and shadow effects of each tile, the light and shadow effect of one tile may be a highlight effect, and the light and shadow effect of another tile may be a shadow effect. By obtaining text description information for multiple tiles separately, the generated target icon can be made more realistic and have richer details.
[0036] Specifically, the text description information corresponding to any tile may include but is not limited to at least one of the following: the number, category, position, size, material, shape, color, brightness, transparency and dynamic effect of the target elements in the tile. Still taking the target icon as a wood icon as an example, assuming that the target element is the light and shadow effect on the wood icon, the text description information may include the number of light and shadow effects in the tile. For example, a number of 0 means that no light and shadow effect is added to the tile; a number of 1 means that a light and shadow effect is added to the tile. The text description information may also include the category of light and shadow effects (such as highlight effects or shadow effects), the location of light and shadow effects (for example, adding light and shadow effects to the upper, middle or lower part of the tile to simulate the effect of light exposure from different directions). Assuming that the target element is a color gradient effect on a wood icon, the text description information may include the color, brightness, transparency of the gradient color in the tile, and the location where the color gradient occurs. Assuming that the target element is a shape on the wood icon (such as a flower-shaped element), the text description information can describe the number of flower-shaped elements, the position of the element in the tile, the material (such as real flower material, plastic material, silk material, etc.), the shape (such as round petals or long petals), the color (such as red or yellow) and the dynamic effect, etc.
[0037] In step S16, both the input image and the text description information may be input into a pre-trained target image generation model so that the target image generation model generates an output image, wherein the output image includes a target icon having the morphological features described by the input image and including the target element described by the text description information. In some embodiments, the input image may be an existing image including a target icon. Using the method of the embodiments of this specification, the existing target icon may be customized and updated using the text description information, thereby improving the personalization and customization of the target icon. Figure 2 The output image corresponding to the input image shown is as follows Figure 4 As shown. Figure 4 It can be seen that, through the processing of the embodiment of the present disclosure, the target icon in the output image has more details, thereby presenting a more delicate and realistic visual effect.
[0038] Further, in an embodiment where the input image includes a plurality of tiles, the output image generated by the target image generation model may also include a plurality of tiles, and the plurality of tiles in the output image correspond one-to-one to the plurality of tiles in the input image. The target image generation model may determine each tile in the input image by edge recognition of the input image. In the case where the colors of the plurality of tiles in the input image are different, each tile in the input image may also be determined based on the colors of the plurality of tiles in the input image. In addition, the target image generation model may also determine each tile in the input image by semantic recognition or other methods, which are not listed here one by one. In the case where the text description information includes text description information corresponding to at least one tile in the plurality of tiles in the input image, the target image generation model may process the at least one tile in the input image based on the text description information corresponding to the at least one tile in the input image, and obtain the at least one tile in the output image. For example, the text description information may be "adding a color gradient effect from red to blue to the top tile in the input image", "adding a shadow effect to the red tile in the input image", etc. According to the text description information, the tile in the output image may have a color gradient effect or a shadow effect.
[0039] Furthermore, in an embodiment where the input image is also used to describe color features of a target icon, the output image includes a target icon having morphological features and color features described by the input image and including a target element described by the text description information.
[0040] In some embodiments, the text description information can also describe the color features of the target icon. Further, when both the input image and the text description information can describe the color features of the target icon, there may be a conflict between the color features described by the input image and the color features described by the text description information. For example, the color feature described by the input image is red, while the color feature described by the text description information is blue. In order to reduce the above conflict, different priorities can be set for the color features described by the input image and the color features described by the text description information. The color feature with a higher priority between the color feature described by the input image and the color feature described by the text description information can be determined as the target color feature, and an output image of the target icon including the target color feature can be generated by the target image generation model. Still assuming that the color feature described by the input image is red, and the color feature described by the text description information is blue, when the color feature described by the text description information has a higher priority, blue can be used as the target color feature, so that the target image generation model generates an output image including a blue icon.
[0041] In some embodiments, the target image generation model can be trained based on the input sample image and the sample text description information corresponding to the input sample image. The input sample image can be obtained from an image database, and the sample text description information corresponding to the input sample image can be generated by manual annotation or other means. Specifically, the output sample image generated by the initial image generation model (i.e., the untrained image generation model) based on the sample text description information can be obtained, and based on the similarity between the output sample image and the input sample image, and the similarity between the feature vector corresponding to the output sample image and the feature vector corresponding to the text description information, the initial image generation model is trained to obtain the target image generation model.
[0042] Furthermore, since different icon categories have different characteristics, multiple candidate image generation models can be pre-trained, and the multiple candidate image generation models correspond to multiple icon categories. In actual application, the target icon category corresponding to the target icon can be obtained, and the candidate image generation model corresponding to the target icon category among the multiple candidate image generation models can be determined as the target image generation model. Figure 3 In order to facilitate user selection, options 30 corresponding to each icon category can be displayed on the user interface 10, and the icon category corresponding to the option 30 selected by the user is determined as the target icon category. Assuming that the icon categories include wood category, gem category and equipment category, the candidate image generation model corresponding to the wood category, the candidate image generation model corresponding to the gem category and the candidate image generation model corresponding to the equipment category can be generated respectively. Assuming that the target icon category corresponding to the target icon is the gem category, the candidate image generation model corresponding to the gem category can be determined as the target image generation model. After determining the target image generation model, the user can upload an input image of the target icon including the target icon category, so that the target image generation model generates an output image based on the input image and text description information uploaded by the user.
[0043] In other embodiments, the user may upload an input image on the user interface 10, and then provide one or more category options on the user interface 10, from which the user may select one as a target category option, and the icon category corresponding to the target category option may be determined as the target icon category. Then, the candidate image generation model corresponding to the target icon category may be determined as the target image generation model, and the output image may be generated by the target image generation model based on the input image and text description information uploaded by the user.
[0044] Alternatively, in other embodiments, options 30 corresponding to each icon category may be displayed on the user interface 10, the icon category corresponding to the option 30 selected by the user may be determined as the target icon category, the candidate image generation model corresponding to the target icon category may be determined as the target image generation model, and candidate input images corresponding to the target category options may be displayed on the user interface 10 (which may be automatically provided by the system, for example, obtained from a pre-established image data block), the user may select an input image from the candidate input images, and an output image may be generated by the target image generation model based on the input image selected by the user and the text description information.
[0045] In some embodiments, after the output image is acquired, the style of the target icon in the output image may be transferred using a pre-trained target style transfer model, thereby adjusting the target icon to a style expected by the user.
[0046] Among them, the target style transfer model can be selected from multiple candidate style transfer models. In some embodiments, the above-mentioned multiple candidate style transfer models correspond to multiple candidate styles one by one. The target style determined by the user from multiple candidate styles can be obtained, and the candidate style transfer model corresponding to the target style is determined as the target style transfer model. In this embodiment, the user can flexibly select the target style according to his own preferences, so as to convert the target icon into a target style that matches the user's preferences, and realize the style customization of the target icon according to the user's preferences. In other embodiments, multiple candidate style transfer models correspond to multiple application scenarios one by one. The target application scenario determined by the user from multiple application scenarios can be obtained, and the candidate style transfer model corresponding to the target application scenario is determined as the target style transfer model. Among them, the application scenario can be selected by the user, or obtained as an input parameter from the parameter transmission interface. For example, in a game scene, the style of the game icon can be adapted to the style of the game. Assuming that the style of the game is a cartoon style, the target icon can be converted into a cartoon style. For another example, in a cultural education scene, an ink-and-wash style icon can be used to display Chinese traditional culture and art, while an oil painting style can be used to introduce Western art history. Therefore, the target icon can be converted into ink style or oil painting style according to actual needs.
[0047] The style transfer model can be trained based on the input sample images uploaded by the user. The trained style transfer model can transfer the output image to the same style as the input sample images uploaded by the user. For example, if the style of the input sample image is oil painting style, the trained style transfer model can transfer the output image to oil painting style.
[0048] In some embodiments, user feedback on the output image may also be obtained, and the target image generation model may be adjusted based on the feedback information. The feedback information may be a score, and a higher score indicates that the user is more satisfied with the output image generated by the target image generation model. Figure 3 , the output image can be displayed on the user interface 10, and a rating option can be provided. The user can preview the output image in real time on the user interface 10, and give the rating corresponding to the output image through the rating option 40 (the rating method shown in the figure is a star rating, and it can be understood that other rating forms can also be used, such as a percentage rating). Through the above method, the interactivity and efficiency in the icon design process can be improved, and it is convenient for users to quickly modify icons according to their needs. Alternatively, the feedback information can also be category information, such as "good review", "medium review", "bad review", etc. Alternatively, the feedback information can also include evaluation information on at least one dimension of the output image (for example, aesthetic quality dimension, color tone dimension, detail richness dimension, etc.). For example, the feedback information can be the following types of information: "The icon is not good-looking enough", "The details of the icon are not rich enough", etc. If the feedback information indicates that the user is less satisfied with the output image (for example, the user's rating of the output image is lower than the preset rating threshold), the target image generation model can be adjusted.
[0049] The generated output image can be used in a variety of different projects. For example, in a game scene, the target icon can be a game icon, and the output image can be directly imported into game development tools and game engines (such as Unity, Unreal Engine), so that the target icon in the output image can be directly applied to the game project, simplifying the workflow.
[0050] The embodiments of this specification have the following advantages:
[0051] (1) Icons are automatically generated through the target image generation model, which reduces the time and effort of designers in manual drawing, allowing the development team to complete projects more quickly and shorten the project development cycle.
[0052] (2) Automated icon generation reduces the reliance on a large number of art designers, thereby reducing labor costs.
[0053] (3) The target image generation model can learn and generate icons of various styles from a huge image database, providing a rich source of creativity and helping designers explore new design directions.
[0054] (4) Developers can generate personalized icons based on different themes and styles by adjusting parameters (for example, the pixel value of the target icon in the input image or the specific content of the text description information).
[0055] (5) After the icon is generated, user feedback can be quickly obtained. Developers can make adjustments and optimizations based on the feedback to improve the user experience.
[0056] An embodiment of the present specification also provides a computer device, which at least includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any of the above embodiments is implemented.
[0057] Figure 5 A more specific hardware structure diagram of a computer device provided in an embodiment of this specification is shown, and the device may include: a processor 202, a memory 204, an input / output interface 206, a communication interface 208, and a bus 210. The processor 202, the memory 204, the input / output interface 206, and the communication interface 208 are connected to each other through the bus 210 in the device.
[0058] The processor 202 can be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing relevant programs to implement the technical solutions provided in the embodiments of this specification. The processor 202 can also include a graphics card, which can be a Nvidia titan X graphics card or a 1080Ti graphics card, etc.
[0059] The memory 204 can be implemented in the form of a read-only memory (ROM), a random access memory (RAM), a static storage device, a dynamic storage device, etc. The memory 204 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 204 and are called and executed by the processor 202.
[0060] The input / output interface 206 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.
[0061] The communication interface 208 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0062] The bus 210 comprises a pathway for transmitting information between the various components of the device, such as the processor 202 , the memory 204 , the input / output interface 206 , and the communication interface 208 .
[0063] It should be noted that, although the above device only shows the processor 202, the memory 204, the input / output interface 206, the communication interface 208 and the bus 210, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.
[0064] An embodiment of the present specification provides a computer program product, including a computer program. When the computer program is executed by a processor, the method described in any embodiment of the present specification is implemented.
[0065] The embodiments of the present specification also provide a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any of the above embodiments is implemented.
[0066] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computer device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0067] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is merely schematic, wherein the modules described as separate components may or may not be physically separated, and the functions of each module can be implemented in the same one or more software and / or hardware when implementing the embodiment scheme of this specification. It is also possible to select some or all of the modules according to actual needs to achieve the purpose of the embodiment scheme. A person of ordinary skill in the art can understand and implement it without paying creative labor.
[0068] The above is only a specific implementation of the embodiments of this specification. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the embodiments of this specification. These improvements and modifications should also be regarded as the protection scope of the embodiments of this specification.
Claims
1. A method for generating an icon, the method comprising: Acquire an input image including a target icon, wherein the input image is used to describe morphological features of the target icon; Acquire text description information corresponding to the target icon, where the text description information is used to describe the target element on the target icon; The input image and the text description information are processed by a pre-trained target image generation model to obtain an output image, wherein the output image includes the target icon having the morphological features and including the target element. 2 . The method according to claim 1 , wherein the target icon comprises a plurality of tiles, and the text description information comprises text description information corresponding to at least one tile among the plurality of tiles.
3. According to the method of claim 2, the text description information corresponding to any block includes at least one of the following: the number, category, position, size, material, shape, color, brightness, transparency and dynamic effect of the target elements in the block. 4 . According to the method of claim 1 , the input image is also used to describe the color features of the target icon, and the output image includes the target icon having the morphological features and the color features and including the target element.
5. The method according to claim 4, wherein the text description information is further used to describe the color characteristics of the target icon; the output image includes the target icon having the morphological characteristics and the target color characteristics and including the target element, wherein: The target color feature is a color feature with a higher priority between the color feature described by the input image and the color feature described by the text description information.
6. The method according to claim 1, wherein the target image generation model is selected from a plurality of candidate image generation models, and the plurality of candidate image generation models correspond to a plurality of icon categories; The method further comprises: Obtaining a target icon category corresponding to the target icon; A candidate image generation model corresponding to the target icon category among the multiple candidate image generation models is determined as the target image generation model.
7. The method according to claim 1, further comprising: The style of the target icon in the output image is transferred using a pre-trained target style transfer model.
8. The method according to claim 7, wherein the target style transfer model is selected from a plurality of candidate style transfer models, and the plurality of candidate style transfer models correspond one-to-one to a plurality of candidate styles; the method further comprising: Obtaining a target style determined by a user from the multiple candidate styles; Determine a candidate style transfer model corresponding to the target style as the target style transfer model; or The target style transfer model is selected from a plurality of candidate style transfer models, and the plurality of candidate style transfer models correspond one-to-one to a plurality of application scenarios; The method further comprises: Acquire a target application scenario determined by a user from the multiple application scenarios; A candidate style transfer model corresponding to the target application scenario is determined as the target style transfer model.
9. The method according to claim 1, further comprising: Obtaining user feedback information on the output image; The target image generation model is adjusted based on the feedback information.
10. A computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method according to any one of claims 1 to 9.
11. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 9 when executing the program.
12. A computer program product, comprising a computer program, which implements the method according to any one of claims 1 to 9 when executed by a processor.
Citation Information
Patent Citations
Style image generation method and device, computer equipment and storage medium
CN116071452A
Icon generation method and device, storage medium and electronic equipment
CN116824004A
Image generation method and device, computer equipment and storage medium
CN117112826A
Icon generation method, electronic equipment and computer storage medium
CN117519881A
Image generation method and device
CN117710510A
Cited By
Image generation method, medium, computer device, and program product
EP4745897A1
