Interactive interface adjusting method and device, electronic equipment and readable storage medium

Abstract

The invention discloses an interactive interface adjusting method and device, electronic equipment and a readable storage medium, and can be applied to the technical field of computer vision. The method comprises the steps that in response to a reconstruction instruction acting on the interactive interface, feature description information of the interactive interface is obtained, and the feature description information is used for describing the interactive interface; the feature description information and the random noise vector are input into a target generation model for analysis, at least one design drawing corresponding to the feature description information is obtained, and the target generation model is obtained by training a deep learning model through interface design element samples and feature description information samples in advance; and in response to a selection operation for the at least one design drawing, selecting a target design drawing from the at least one design drawing, and reconstructing the interactive interface based on the target design drawing. Through the method and the device, the technical problems that a traditional interactive interface is relatively high in complexity and is difficult to match the use habits of the user are solved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method, device, electronic device, and readable storage medium for adjusting an interactive interface. Background Art

[0002] Currently, traditional software interfaces are typically designed by the operating company, using a uniform style and layout. These interfaces often include an increasing number of software features, often requiring users to learn and memorize specific interaction patterns and methods. This leads to a steep learning curve for new users and increases the difficulty of using the product or application. As technology and user expectations evolve, modern design trends emphasize simplicity, flatness, and responsiveness. Traditional interface design can appear overly complex and outdated, inconsistent with modern design trends. Furthermore, traditional interface design may overlook the accessibility needs of certain user groups, such as those with visual impairments, hearing impairments, or physical disabilities. This can prevent some users from effectively using the product or application. Consequently, there is a technical issue with interfaces being complex and difficult to adapt to user habits.

[0003] Currently, no effective solution has been proposed to the technical problem that the interactive interface in related technologies is highly complex and difficult to match with user usage habits. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, electronic device and readable storage medium for adjusting an interactive interface to solve the technical problem that traditional interactive interfaces are highly complex and difficult to match user usage habits.

[0005] To achieve the above-mentioned objectives, according to one aspect of the present application, a method for adjusting an interactive interface is provided. The method comprises: in response to a reconstruction instruction acting on the interactive interface, obtaining feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface; inputting the feature description information and a random noise vector into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples; in response to a selection operation on at least one design drawing, selecting a target design drawing from the at least one design drawing, and reconstructing the interactive interface based on the target design drawing.

[0006] Optionally, before reconstructing the interactive interface based on the target design drawing, the adjustment method of the interactive interface also includes: obtaining adjustment description information in response to an adjustment instruction for the target design drawing; adjusting the target design drawing based on the adjustment description information to obtain an adjusted target design drawing; and updating the target design drawing using the adjusted target design drawing.

[0007] Optionally, reconstructing the interactive interface based on the target design drawing includes: adjusting the style features of the interactive interface based on the style features of the target design drawing; and adjusting the layout structure of the interactive interface based on the layout structure of the target design drawing.

[0008] Optionally, the adjustment method of the interactive interface also includes: in response to the deep learning model being a generative adversarial network model, using interface design element samples and feature description information samples to train the generative adversarial network model to obtain a target generative model.

[0009] Optionally, the generative adversarial network model includes a generator and a discriminator, and the generative adversarial network model is trained using interface design element samples and feature description information samples to obtain a target generative model, including: initializing the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; using the interface design element samples to train the discriminator, wherein, during the training of the discriminator, the model parameters of the generator remain fixed; in response to the completion of the discriminator training, using the interface design element samples and feature description information samples to train the generator, wherein, during the training of the generator, the model parameters of the discriminator remain fixed.

[0010] Optionally, the discriminator is trained using interface design element samples, including: inputting the interface design elements randomly generated by the generator and the interface design element samples into the discriminator for identification to obtain a first identification result, wherein the first identification result is used to evaluate the ability of the discriminator to distinguish between the randomly generated interface design elements and the interface design element samples; in response to the first identification result failing to meet the identification performance index, adjusting the model parameters of the discriminator, wherein the identification performance index is used to represent the correct classification rate of the discriminator for the interface design element samples and the incorrect classification rate of the randomly generated interface design elements; inputting the interface design elements randomly generated by the generator and the interface design element samples into the adjusted discriminator for identification to obtain a second identification result of the discriminator; in response to the second identification result meeting the identification performance index, determining that the discriminator training is complete.

[0011] Optionally, the generator is trained using interface design element samples and feature description information samples, including: inputting the feature description information sample and the random noise vector into the generator to obtain a first predicted interface design element; inputting the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; in response to the first evaluation result not meeting the generation performance standard, adjusting the model parameters of the generator to obtain an adjusted generator; inputting the feature description information sample and the random noise vector into the adjusted generator to obtain a second predicted interface design element; inputting the second predicted interface design element and the interface design element sample into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; in response to the second evaluation result meeting the generation performance index, determining that the generator training is completed.

[0012] To achieve the above-mentioned purpose, according to another aspect of the present application, a device for adjusting an interactive interface is provided. The device includes: an acquisition unit for responding to a reconstruction instruction acting on the interactive interface and acquiring feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface; an input unit for inputting the feature description information and a random noise vector into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples; a reconstruction unit for responding to a selection operation on at least one design drawing, selecting a target design drawing from the at least one design drawing, and reconstructing the interactive interface based on the target design drawing.

[0013] Optionally, before reconstructing the interactive interface based on the target design drawing, the adjustment method of the interactive interface also includes: obtaining adjustment description information in response to an adjustment instruction for the target design drawing; adjusting the target design drawing based on the adjustment description information to obtain an adjusted target design drawing; and updating the target design drawing using the adjusted target design drawing.

[0014] Optionally, the reconstruction unit is further configured to: adjust the style features of the interactive interface based on the style features of the target design drawing; and adjust the layout structure of the interactive interface based on the layout structure of the target design drawing.

[0015] Optionally, the adjustment device of the interactive interface also includes: a training unit, which is used to train the generative adversarial network model using interface design element samples and feature description information samples in response to the deep learning model being a generative adversarial network model to obtain a target generative model.

[0016] Optionally, the training unit is also used to: initialize the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; train the discriminator using interface design element samples, wherein, during the training of the discriminator, the model parameters of the generator remain fixed; in response to the completion of the discriminator training, train the generator using interface design element samples and feature description information samples, wherein, during the training of the generator, the model parameters of the discriminator remain fixed.

[0017] Optionally, the training unit is also used to: input the interface design elements and interface design element samples randomly generated by the generator into the discriminator for recognition to obtain a first recognition result, wherein the first recognition result is used to evaluate the discriminator's ability to distinguish between randomly generated interface design elements and interface design element samples; in response to the recognition result failing to meet the recognition performance index, adjust the model parameters of the discriminator, wherein the recognition performance index is used to represent the discriminator's correct classification rate of the interface design element samples and the incorrect classification rate of the randomly generated interface design elements; input the interface design elements and interface design element samples randomly generated by the generator into the adjusted discriminator for recognition to obtain a second recognition result of the discriminator; in response to the second recognition result meeting the recognition performance index, determine that the discriminator training is completed.

[0018] Optionally, the training unit is also used to: input the feature description information sample and the random noise vector into the generator to obtain a first predicted interface design element; input the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; in response to the first evaluation result not meeting the generation performance standard, adjust the model parameters of the generator to obtain an adjusted generator; input the feature description information sample and the random noise vector into the adjusted generator to obtain a second predicted interface design element; input the second predicted interface design element and the interface design element sample into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; in response to the second evaluation result meeting the generation performance index, determine that the generator training is completed.

[0019] To achieve the above object, according to another aspect of the present application, an electronic device is provided, which includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0020] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the steps in any one of the above-mentioned method embodiments.

[0021] In order to achieve the above object, according to another aspect of the present application, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the computer program implements the steps in any of the above method embodiments.

[0022] In an embodiment of the present application, in response to a reconstruction instruction acting on an interactive interface, feature description information of the interactive interface is obtained, wherein the feature description information is used to describe the interactive interface; the feature description information and a random noise vector are input into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples; in response to a selection operation for at least one design drawing, a target design drawing is selected from at least one design drawing, and the interactive interface is reconstructed based on the target design drawing. That is, in an embodiment of the present application, by responding to the user's reconstruction instruction and feature description information, at least one design drawing that matches the user's preferences and needs can be generated, and then the interactive interface can be adjusted according to the target design drawing selected by the user, so that the interactive interface can be customized according to the user's usage habits to generate a design drawing that satisfies the user, improve the user experience, and achieve the technical effect of a customized interactive interface, thereby solving the technical problem that the traditional interactive interface is highly complex and difficult to match the user's usage habits. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0024] Figure 1 is a hardware structure block diagram of a computer terminal for implementing a method for adjusting an interactive interface according to an embodiment of the present application;

[0025] Figure 2 is a flow chart of a method for adjusting an interactive interface according to an embodiment of the present application;

[0026] Figure 3 is a schematic diagram of an adjustment device for an interactive interface according to an embodiment of the present application;

[0027] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] It should be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation portals for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation portals for users to choose to agree or refuse the automated decision-making results; if the user chooses to refuse, the expert decision-making process will be entered.

[0031] Example 1

[0032] According to an embodiment of the present application, an embodiment of a method for adjusting an interactive interface is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1This is a hardware structure block diagram of a computer terminal (or mobile device) for implementing an adjustment method for an interactive interface according to an embodiment of the present application. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (illustrated as 102a, 102b, ..., 102n in the figure) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Micro controller Unit, abbreviated as MCU) or a programmable logic device (Field-Programmable Gate Array, abbreviated as FPGA)), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (Input / Output, abbreviated as I / O), a universal serial bus (Universal Serial Bus, abbreviated as USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0034] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0035] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for adjusting the interactive interface in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the above-mentioned method for adjusting the interactive interface. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0036] The transmission device 106 is used to receive or send data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0037] The display may be, for example, a touch screen liquid crystal display (LCD), which enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0038] Under the above operating environment, this application provides Figure 2 The adjustment method of the interactive interface shown. Figure 2 This is a flowchart of the method for adjusting the interactive interface according to Example 1 of the present application.

[0039] Step S201 : In response to a reconstruction instruction acting on an interactive interface, feature description information of the interactive interface is obtained.

[0040] In the technical solution provided in step S201 above, the interactive interface can be a software (Application, referred to as APP) interface, for example, the interface of a mobile banking APP. The reconstruction instruction can be a voice instruction or a text instruction used to trigger the reconstruction of the interactive interface. The feature description information is used to describe the interactive interface.

[0041] In this embodiment, the interactive interface may be provided with a "Customize" component or an "Interface Design" component. The user may trigger a refactoring instruction and initiate the refactoring process by clicking the "Customize" component or the "Interface Design" button on the interface. Subsequently, the user (target object) obtains feature description information for the interactive interface. The feature description information may be in voice or text form, and the user may use the feature description information to express specific requirements for adjusting various elements of the interactive interface. For example, the theme, style, color, layout style, etc. of the interactive interface may be selected.

[0042] For example, after triggering a refactoring command, a user might enter a feature description like "The theme is ocean-themed, with primarily blue and green colors. The layout places the search box at the top of the screen, and the menu button at the bottom." This feature description encompasses requirements across multiple dimensions, and the system needs to be able to understand and parse this information, converting it into a format that the machine learning model can recognize and process, and then generate an interface design that conforms to the user's feature description. Users can also adjust font size, icon style, button position, and other features as required to meet different usage scenarios and needs.

[0043] In this step, if the user is dissatisfied with the interface, they can trigger a refactoring command and enter a feature description of the interface to express their need for adjustments. The system can then flexibly respond to the user's feature description to personalize the current interface. By allowing the user to directly input feature descriptions, the system can more accurately capture user preferences and refactor the interface to meet their needs and aesthetics. This avoids the "one-size-fits-all" problem of traditional design, allowing each user to have an interface that better suits their individual needs and preferences.

[0044] Step S202: input the feature description information and the random noise vector into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information.

[0045] In the technical solution provided in step S202 above, the target generation model is obtained by pre-training a deep learning model using samples of interface design elements and feature description information. Upon receiving the feature description information provided by the user, the system can input the feature description information and a random noise vector into the target generation model for analysis, thereby obtaining at least one design drawing corresponding to the feature description information.

[0046] In this embodiment, after receiving the processed feature description information and random noise vector, the target generation model begins analysis and generation. The target generation model uses the feature description information as conditional input and the random noise vector as a random seed for generation. Through data processing and algorithmic calculations, the target generation model can generate at least one design that matches the user-provided feature description information in terms of style, color, layout, and other aspects.

[0047] Optionally, in addition to direct user input, feature descriptions can be supplemented and refined by combining historical user data, behavioral preferences, and recent contextual information. For example, when a user says, "I want to turn the interface red," recommendations can be made based on the red elements contained in the user's browsing data over a short period of time.

[0048] In this step, the feature description information input by the user can be analyzed by a pre-trained target generation model, and then at least one design drawing matching the feature description information can be generated, so as to achieve the purpose of providing diversified personalized interface design solutions.

[0049] Step S203 : In response to the selection operation on at least one design drawing, a target design drawing is selected from the at least one design drawing, and the interactive interface is reconstructed based on the target design drawing.

[0050] In the technical solution provided in step S203 above, after at least one design drawing is designed in step S202, the at least one generated design drawing can be displayed to the user, who can observe these designs and select the one that best suits their personal aesthetic. The selection operation can be a click operation, a drag selection on the touch screen, or a voice command confirmation, without specific restrictions here. Through this link, users can directly participate in design decisions, enhancing user autonomy and satisfaction.

[0051] In this embodiment, after the user selects a target design from at least one design, the interactive interface can be reconstructed according to the target design, wherein the target design includes all elements and features that the user wants to be reflected in the reconstructed interactive interface.

[0052] Optionally, the system begins refactoring the existing interface based on the target design. This can include rearranging interface elements, changing colors and styles, adjusting icons and fonts, and so on. This refactoring process ensures that the new interface design not only remains visually consistent with the target design but also maintains usability, responsiveness, and functionality.

[0053] Optionally, after the interactive interface is reconstructed, the system will show it to the user again for final confirmation. Once the user is satisfied, the system will save this design as the user's default interactive interface, allowing the user to see and use the personalized interface design selected by the user in subsequent uses.

[0054] In the above steps S201 to S203, by responding to the user's reconstruction instructions and feature description information, at least one design drawing that matches the user's preferences and needs can be generated, and then the interactive interface can be adjusted according to the target design drawing selected by the user, so that the interactive interface can be customized according to the user's usage habits to generate a design drawing that satisfies the user, improve the user experience, and achieve the technical effect of customized interactive interface, thereby solving the technical problem that traditional interactive interfaces are highly complex and difficult to match user usage habits.

[0055] The embodiment of the present invention is described in detail below in conjunction with the above steps.

[0056] As an optional implementation, before reconstructing the interactive interface based on the target design drawing, the adjustment method of the interactive interface also includes: obtaining adjustment description information in response to an adjustment instruction for the target design drawing; adjusting the target design drawing based on the adjustment description information to obtain an adjusted target design drawing; and updating the target design drawing using the adjusted target design drawing.

[0057] In this embodiment, after the target design drawing is determined, the user is also supported to adjust the target design drawing so that the adjusted target design drawing better meets the user's expectations.

[0058] For example, if the user wants to change the color of a specific button on the target design, adjust the icon size or position, modify the font style, etc., the user can issue adjustment instructions through the interactive interface. These instructions can be direct click operations, drag adjustments, or more detailed text descriptions, which are not specifically limited here.

[0059] Optionally, in response to the user's adjustment instructions, the system will further collect adjustment description information. This adjustment description information may include specific design elements, desired style changes, color adjustments, size changes, etc. In other words, the adjustment description is the user's specific requirements for fine-tuning the target design. The system can recognize and understand this adjustment description information and convert it into a machine-processable format to prepare for the next adjustment.

[0060] Optionally, after receiving the adjustment description information provided by the user, the system will adjust the target design drawing based on this information. For example, using the adjustment description information as an input condition, the selected design drawing is partially or completely modified to be closer to the user's final needs. The system may also provide an intuitive adjustment tool that allows users to operate directly on the target design drawing. For example, adjust the color saturation through a slider or change the position of an element by dragging. After adjusting the target design drawing, the adjusted target design drawing can be obtained; and then the target design drawing is updated using the adjusted target design drawing, so that the updated target design drawing better meets user needs.

[0061] In this step, the user is allowed to adjust the target design drawing, so that the adjusted target design drawing can better meet the user's needs, and then the interactive interface can be reconstructed using the updated target design drawing, so that the reconstructed interactive interface can accurately match the user's personalized requirements, thereby providing a better user experience.

[0062] As an optional implementation, step S203 reconstructs the interactive interface based on the target design drawing, including: adjusting the style features of the interactive interface based on the style features of the target design drawing; adjusting the layout structure of the interactive interface based on the layout structure of the target design drawing.

[0063] In this embodiment, after the target design drawing is determined, the interactive interface can be adjusted according to the style characteristics and layout structure of the target design drawing so that the adjusted interactive interface meets the needs of the user.

[0064] For example, the system can extract the stylistic features of the target design and then apply them to the interactive interface to adjust the style of the interactive interface. For example, if the stylistic features of the target design are "cyberpunk," the system will attempt to incorporate the colors (such as neon effects), icons (such as futuristic symbols), and button styles (such as metallic textures) of this style into the interactive interface to achieve visual consistency.

[0065] Optionally, the layout features of the target design drawing are analyzed, for example, the relative position and size of each user interface (UI) element (such as menus, input boxes, buttons) in the design drawing are parsed, and then the parsed layout structure is applied to the current interactive interface to adjust the position, size and arrangement of the elements.

[0066] Optionally, after completing the style and layout adjustments based on the target design, the adjusted interface can be displayed for user confirmation. The user can check whether the adjusted interface meets their personal needs and achieves the expected effect in terms of style and layout. If the user is satisfied with the adjusted interface, the system will save it as the user's personalized settings. If the user has further modification suggestions, the system will adjust the interface again based on the user's feedback until the user is completely satisfied.

[0067] In this step, the style features and layout structure of the interactive interface are adjusted according to the style features and layout structure of the target design drawing, which can achieve direct conversion from user needs to UI design drawings, and then accurate reconstruction from design drawings to interactive interfaces, ensuring the complete realization of user personalized needs, overcoming the limitations of traditional UI interface design, and providing users with a more flexible, personalized and efficient interactive interface.

[0068] As an optional implementation, the method for adjusting the interactive interface also includes: in response to the deep learning model being a generative adversarial network model, using interface design element samples and feature description information samples to train the generative adversarial network model to obtain a target generation model.

[0069] In this embodiment, the deep learning model can be a generative adversarial network or an autoencoder. The generative adversarial network is used as an example here to collect a large number of interface design element samples and feature description information samples. The interface design element samples can be UI design screenshots, layout diagrams, style guides, etc. of various applications and websites. These samples contain rich UI design elements, such as buttons, icons, menus, color schemes, etc. The feature description information samples are descriptions of various user requirements for UI design, which can be text, voice, or user behavior data, such as preference settings, commonly used functions, etc. These data samples will be used to train the generative adversarial network model to enable it to understand the characteristics of different UI design elements and the diversity of user needs.

[0070] Optionally, after obtaining the interface design element samples and the feature description information samples, the generative adversarial network model may be trained based on the interface design element samples and the feature description information samples to obtain a target generative model.

[0071] In this step, the generative adversarial network model can be trained based on the interface design element samples and feature description information samples to obtain a target generation model, so that the subsequent target generation model can generate realistic UI design drawings that meet specific requirements based on the user's feature description information.

[0072] As an optional implementation, the generative adversarial network model includes a generator and a discriminator. The generative adversarial network model is trained using interface design element samples and feature description information samples to obtain a target generative model, including: initializing the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; using the interface design element samples to train the discriminator, wherein, during the training of the discriminator, the model parameters of the generator remain fixed; in response to the completion of the discriminator training, using the interface design element samples and feature description information samples to train the generator, wherein, during the training of the generator, the model parameters of the discriminator remain fixed.

[0073] In this embodiment, the generative adversarial network model consists of two main components: a generator and a discriminator. The generator is used to generate new UI designs based on input feature description information and random noise; the discriminator is used to determine whether the generated UI designs are realistic and indistinguishable from the real ones.

[0074] Optionally, when training a generative adversarial network model using samples of interface design elements and feature description information, the model parameters of the generator and the discriminator can be initialized. Model parameters include weight parameters and bias parameters, and are typically initialized using random initialization.

[0075] Optionally, the discriminator can be trained first so that it can accurately distinguish between real design drawings and fake design drawings generated by the generator. At this stage, the model parameters of the generator remain fixed and do not participate in back-propagation updates. The system alternately extracts data from real UI design element samples and interface design elements randomly generated by the generator, and inputs them into the discriminator for discrimination. The goal of the discriminator is to label real samples as 1 and randomly generated samples as 0. The discriminator measures its classification performance by calculating a loss function (such as binary cross entropy). The smaller the value of the loss function, the better the performance of the discriminator. Through back-propagation and gradient descent, the model parameters of the discriminator are updated so that it can more accurately distinguish between real and fake samples.

[0076] Optionally, after the discriminator has been trained to meet the requirements, the generator can be trained. During the generator training process, the discriminator model parameters remain fixed. The generator generates UI designs based on user feature description information samples. By calculating the loss function (usually the loss associated with the discriminator) and performing backpropagation, the generator model parameters are updated to optimize the generator's generation capabilities.

[0077] Optionally, training a GAN is an iterative process, with the generator and discriminator training in turn until convergence is achieved or a pre-set training goal is met. In each training iteration, the generator is fixed and the discriminator is trained, then the discriminator is fixed and the generator is trained again, alternating until the generator can consistently produce high-quality UI designs that the discriminator cannot distinguish from real designs.

[0078] In this step, when the GAN model training achieves the desired effect, the generator in the GAN model can be used as the target generation model. At this point, the target generator can generate customized UI designs based on different input feature descriptions (such as user preferences and functional requirements), thereby improving the user experience.

[0079] As an optional implementation, the discriminator is trained using interface design element samples, including: inputting the interface design elements randomly generated by the generator and the interface design element samples into the discriminator for identification to obtain a first identification result, wherein the first identification result is used to evaluate the ability of the discriminator to distinguish between the randomly generated interface design elements and the interface design element samples; in response to the first identification result failing to meet the identification performance index, adjusting the model parameters of the discriminator, wherein the identification performance index is used to represent the correct classification rate of the discriminator for the interface design element samples and the incorrect classification rate for the randomly generated interface design elements; inputting the interface design elements randomly generated by the generator and the interface design element samples into the adjusted discriminator for identification to obtain a second identification result of the discriminator; in response to the second identification result meeting the identification performance index, determining that the discriminator training is complete.

[0080] In this embodiment, the discriminator can be trained using interface design element samples. In the initial stage of training, the generator randomly generates interface design elements, which are usually not very similar to real samples. The elements generated by the generator ("fake" samples) are input into the discriminator together with the real UI design element samples ("real" samples). The discriminator attempts to identify and distinguish "real" samples from "fake" samples, and outputs a first recognition result. The first recognition result is used to evaluate whether the recognition performance of the discriminator has reached the set recognition performance index. The recognition performance index generally reflects the correct classification rate of the discriminator for "real" samples (that is, the probability of identifying the sample as a real design element sample) and the misclassification rate of false samples (that is, the probability of misidentifying the generated sample as a real design element sample).

[0081] Optionally, if the first recognition result fails to meet the recognition performance index, that is, the discriminator's classification ability for real samples is insufficient or there are too many classification errors for fake samples, then it is necessary to further adjust the model parameters of the discriminator to optimize its recognition ability. After adjusting the model parameters of the discriminator, the interface design element samples randomly generated by the generator and the real design element samples are re-input into the discriminator. The discriminator performs re-recognition and outputs a second recognition result. If the second recognition result meets or exceeds the recognition performance index, it indicates that the discriminator performs well in distinguishing between real design element samples and randomly generated UI design element samples. In this case, the discriminator training can be considered complete.

[0082] In this step, through continuous iteration, the discriminator's model parameters are adjusted and optimized, enabling it to accurately distinguish between real design element samples and randomly generated ones. This in turn promotes the improvement of the generator, allowing it to produce UI elements that are more consistent with real design characteristics. This interactive training mechanism enables the generator to continuously learn, ultimately generating highly realistic and personalized design elements to meet the diverse needs and preferences of users.

[0083] As an optional implementation, the generator is trained using interface design element samples and feature description information samples, including: inputting the feature description information sample and the random noise vector into the generator to obtain a first predicted interface design element; inputting the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; in response to the first evaluation result not meeting the generation performance standard, adjusting the model parameters of the generator to obtain an adjusted generator; inputting the feature description information sample and the random noise vector into the adjusted generator to obtain a second predicted interface design element; inputting the second predicted interface design element and the interface design element sample into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; in response to the second evaluation result meeting the generation performance index, determining that the generator training is complete.

[0084] In this embodiment, the generator is trained to optimize its ability to generate UI design elements through interaction with the discriminator, so that it can generate interface elements that are highly similar to real design element samples based on feature description information and random noise.

[0085] Optionally, the feature description information sample and the random noise vector can be input into the generator to obtain the first predicted interface design element, wherein the random noise vector is used to introduce randomness into the generation process, which helps the generator to generate design element samples with diversity.

[0086] Optionally, after obtaining the first predicted interface design element, the first predicted interface design element produced by the generator and the interface design element sample corresponding to the feature description information sample are input into the trained discriminator for evaluation to obtain a first evaluation result, which reflects the similarity between the first predicted interface design element and the real design element sample.

[0087] Optionally, in response to the first evaluation result failing to meet the generation performance criteria, the weight and bias parameters of the generator can be adjusted based on the feedback from the discriminator through a backpropagation algorithm to improve its generation capability, enabling it to generate elements closer to the actual design. Using the adjusted generator, element generation is performed again: the feature description information sample and the random noise vector are input into the adjusted generator to obtain a second predicted interface design element.

[0088] Optionally, after obtaining the second predicted interface design element, the second predicted interface design element can be input into the trained discriminator for evaluation to obtain a second evaluation result. The second evaluation result reflects the similarity between the design elements produced by the adjusted generator and the real samples. If the similarity reaches or exceeds the predetermined generation performance index, the generator training can be considered complete. Among them, the generation performance index is a standard for measuring the quality of the design elements generated by the generator, including the visual effect of the generated elements, the degree of matching with the feature description information, etc. Achieving this indicator means that the generator can generate high-quality personalized UI design elements based on user needs.

[0089] In this step, the generator gradually learns how to generate interface design elements that are highly similar to real design elements based on the user's feature description information, thereby achieving automated generation of personalized UI designs. This process not only improves the efficiency and flexibility of UI design, but also better meets the personalized needs of users and enhances the user experience.

[0090] The following describes in detail another optional specific implementation.

[0091] Currently, software developers are responsible for designing the interfaces of applications (APPs), forcing users to accept a limited set of themes. On the one hand, user preferences vary, making it difficult to find a single theme that perfectly suits everyone's tastes. On the other hand, software features are becoming increasingly numerous and complex, while users rarely need them. This results in complex and unwieldy software with a high barrier to entry. When designing an interface, designers must balance numerous factors, including functionality, aesthetics, learning curve, and practicality. This results in varying user experiences for different user groups, making it difficult to adapt to a diverse user base.

[0092] In existing technologies, traditional UI design often relies on established and used patterns and layouts. This conservative design approach can lead to a lack of innovation and novelty in the user experience. Traditional UI design is often based on fixed layouts and design elements, making it difficult to adapt to different device types, screen sizes, and resolutions. This can lead to inconsistent user experiences across different platforms and devices. User expectations and demands for interfaces are increasingly diverse, and traditional UI design often fails to fully meet the needs of different user groups. The lack of personalized and customized design elements can lead to dissatisfaction among some users. Traditional UI design often requires users to learn and memorize specific interaction patterns and operating methods. This can lead to a steep learning curve for new users and increase the difficulty of using products or applications. As technology and user expectations continue to evolve, modern design trends emphasize minimalism, flat design, and responsive design. Traditional UI design can appear overly complex and outdated, inconsistent with modern design trends. Traditional UI design can also overlook the accessibility needs of certain user groups, such as those with visual impairments, hearing impairments, or physical disabilities. This can prevent some users from effectively using products or applications. Technical issues arise from the high complexity of UI interfaces, which can be difficult to adapt to user habits.

[0093] However, an embodiment of the present application proposes a method for reconstructing a UI interface, which pre-trains a generative adversarial network model using large-scale UI design elements and user personalized data to obtain a generative model. Afterwards, in response to the user's reconstruction operation on the UI interface, the user's feature description information of the interactive interface can be input into the generative model to generate at least one design drawing that matches the user's needs. The user can select the most satisfactory design drawing from them, and then reconstruct the UI interface according to the design drawing selected by the user. This allows users to independently adjust the UI interface to adapt to their personal usage habits, greatly improving user satisfaction and usage experience, and thus solving the technical problem that the UI interface is highly complex and difficult to match the user's usage habits.

[0094] Next, the training process of the generation model in the embodiment of the present application is introduced.

[0095] In this embodiment, a large-scale dataset containing various UI design elements is constructed. For example, this data can be collected from existing UI design works, or your own dataset can be created using online resources and templates. Descriptive information such as images, texts, functions, etc. that users have recently learned about is collected by monitoring existing user data, image data related to the target image generation task is collected, and preprocessed, such as resizing, cropping, normalization, etc. The data should be weighted according to the date, giving more weight to the most recently acquired information, so as to achieve the final dynamic data analysis recommendation. UI design images are generated using technologies such as generative adversarial networks (GANs) or variational autoencoders (VAEs). These models can generate new design images by learning samples from the dataset.

[0096] Alternatively, consider a Generative Adversarial Network (GAN), a model consisting of a generator and a discriminator. The generator is responsible for generating new image samples, while the discriminator distinguishes between the generator's images and real images. These two models compete with each other through adversarial training, gradually improving the generator's ability to produce realistic images.

[0097] Optionally, a deep learning framework is used to build a generator network to convert random noise into realistic images. A discriminator network is also built using a deep learning framework to distinguish between the images generated by the generator and real images.

[0098] Optionally, the training process for the generator and discriminator is an adversarial process. The goal of the generator is to generate images that are as realistic as possible, while the goal of the discriminator is to accurately distinguish between generated images and real images. Therefore, binary cross-entropy can be used as a loss function to measure the performance of the generator and discriminator. The generator and discriminator are combined and trained alternately.

[0099] Optionally, initialize the weights and biases of the generator and discriminator. Use random initialization or pretrained weights. Fix the generator and freeze its weights so that they do not participate in backpropagation updates at this stage. Train the discriminator by mixing real and generated images into a training batch, labeling real images as 1 and generated images as 0. Feed this batch into the discriminator, calculate the discriminator's loss for the images, and run backpropagation to update the discriminator weights to reduce its classification error for real and generated images.

[0100] Optionally, fix the discriminator's weights so that they are not updated for a certain period of time. Define a loss function, using the cross-entropy loss function, which encourages the generator to produce more realistic samples. This function measures the extent to which the generator's samples are misclassified by the discriminator. Train the generator to deceive the discriminator as much as possible. Repeat this process until the quality of the generated samples meets expectations or converges to a desired state.

[0101] Optionally, evaluate the generator using evaluation metrics (such as quality of generated images, diversity, etc.) and tune it as needed.

[0102] For example, the quality of generated images is evaluated by calculating the distance between the Gaussian distributions of real images and generated images in the feature space. A lower distance indicates that the generated image is closer to the distribution of real images. By structural similarity index: a traditional indicator for evaluating image quality. It measures the structural similarity of two images, with a value between -1 and 1, and the closer to 1, the better the image quality. Image diversity is evaluated by intra-class distance, that is, by calculating the distance between generated images of the same category. A larger intra-class distance indicates that the generated images have better diversity. Image diversity is evaluated by inter-class distance, that is, by calculating the distance between generated images of different categories. A larger inter-class distance indicates that there are obvious differences between the generated images of different categories.

[0103] Optionally, various tuning methods can be used. For example, different model hyperparameters such as learning rate, batch size, and number of training iterations can be tried to find a better model configuration. Increasing the amount of training data can help improve the generalization ability of the generator and image quality. More complex generator network structures can be used, such as increasing the number of layers and changing the activation function.

[0104] Optionally, after obtaining a generative model, you can use it to generate automatic layouts. Collect a set of sample data with good layouts as a training dataset. Use a generative model (such as a variational autoencoder or generative adversarial network) to learn the distribution of the input data and generate layouts similar to the training dataset. Define an appropriate loss function to measure the difference between the generated layout and the true layout. You can use metrics such as pixel-level loss and structural similarity index to measure the similarity of the layouts.

[0105] For example, the calculation method of the structural similarity index can be expressed by the following formula:

[0106] SSIM(x,y)=[(2*μx*μy+C1)*(2*σxy+C2)] / [(μx^2+μy^2+C1)*(σx^2+σy^2+C2)]

[0107] SSIM represents the structural similarity index. C1 and C2 are two constants, typically set to: C1 = (K1 * L)^2, C2 = (K2 * L)^2, where L represents the dynamic range of pixel values (usually 255, representing the maximum pixel value of a grayscale image). K1 and K2 are constants used to adjust for brightness and contrast, typically K1 = 0.01 and K2 = 0.03. μx and μy represent the mean of the two images; σx^2 and σy^2 represent the variance of the two images; and σxy represents the covariance of the two images.

[0108] Optionally, the value range of SSIM is between [-1, 1]. When the SSIM value is closer to 1, the two images are more similar; when the SSIM value is closer to -1, the two images are less similar; when the SSIM value is close to 0, it means that there is a big difference between the two images.

[0109] Optionally, a generative model is trained using the training dataset, and its parameters are optimized by minimizing a loss function. The trained generative model is then used to generate new layouts. The generative process can be guided by inputting random noise or specifying prior conditions to obtain more specific layouts. These models can learn how to automatically arrange UI elements to achieve the best layout results given design requirements.

[0110] Optionally, keyword information extracted based on user voice commands can be used to accurately recommend UIs. By using machine learning algorithms, a recommendation system can be developed to generate personalized UI design suggestions based on user preferences and needs.

[0111] In this embodiment, the trained generative model can be used to personalize the interactive interface based on the user's actual needs. For example, unnecessary functions can be blocked, commonly used function entrances can be placed in front, and a UI interface that better suits user habits can be provided. This lowers the threshold for software use, conforms to user habits, and adapts to the different needs of different users.

[0112] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0113] Example 2

[0114] The embodiment of the present application also provides an interactive interface adjustment device. It should be noted that the interactive interface adjustment device of the embodiment of the present application can be used to execute the interactive interface adjustment method provided in the embodiment of the present application. The interactive interface adjustment device provided in the embodiment of the present application is introduced below.

[0115] According to an embodiment of the present application, a device for implementing the above-mentioned adjustment method of the interactive interface is also provided. Figure 3 As shown, the interactive interface adjustment device 300 includes: an acquisition unit 301 , an input unit 302 and a reconstruction unit 303 .

[0116] The acquisition unit 301 is configured to respond to a reconstruction instruction acting on the interactive interface and acquire feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface.

[0117] The input unit 302 is used to input the feature description information and the random noise vector into the target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples.

[0118] The reconstruction unit 303 is configured to select a target design from the at least one design in response to a selection operation on the at least one design, and reconstruct the interactive interface based on the target design.

[0119] Optionally, the adjustment device of the interactive interface is also used to: obtain adjustment description information in response to an adjustment instruction for the target design drawing; adjust the target design drawing based on the adjustment description information to obtain an adjusted target design drawing; and update the target design drawing using the adjusted target design drawing.

[0120] Optionally, the reconstruction unit 303 is further configured to: adjust the style features of the interactive interface based on the style features of the target design drawing; and adjust the layout structure of the interactive interface based on the layout structure of the target design drawing.

[0121] Optionally, the adjustment device 300 of the interactive interface also includes: a training unit, which is used to train the generative adversarial network model using interface design element samples and feature description information samples in response to the deep learning model being a generative adversarial network model to obtain a target generative model.

[0122] Optionally, the training unit is also used to: initialize the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; train the discriminator using interface design element samples, wherein, during the training of the discriminator, the model parameters of the generator remain fixed; in response to the completion of the discriminator training, train the generator using interface design element samples and feature description information samples, wherein, during the training of the generator, the model parameters of the discriminator remain fixed.

[0123] Optionally, the training unit is also used to: input the interface design elements and interface design element samples randomly generated by the generator into the discriminator for recognition to obtain a first recognition result, wherein the first recognition result is used to evaluate the discriminator's ability to distinguish between randomly generated interface design elements and interface design element samples; in response to the recognition result failing to meet the recognition performance index, adjust the model parameters of the discriminator, wherein the recognition performance index is used to represent the discriminator's correct classification rate of the interface design element samples and the incorrect classification rate of the randomly generated interface design elements; input the interface design elements and interface design element samples randomly generated by the generator into the adjusted discriminator for recognition to obtain a second recognition result of the discriminator; in response to the second recognition result meeting the recognition performance index, determine that the discriminator training is completed.

[0124] Optionally, the training unit is further configured to: input the feature description information sample and the random noise vector into the generator to obtain a first predicted interface design element; input the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; and in response to the first evaluation result failing to meet the generation performance standard, adjust the model parameters of the generator to obtain an adjusted generator;

[0125] The feature description information sample and the random noise vector are input into the adjusted generator to obtain a second predicted interface design element; the second predicted interface design element and the interface design element sample are input into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; in response to the second evaluation result reaching the generation performance index, it is determined that the generator training is completed.

[0126] The interactive interface adjustment device provided in the embodiment of the present application can generate at least one design drawing that matches the user's preferences and needs by responding to the user's reconstruction instructions and feature description information, and then adjust the interactive interface according to the target design drawing selected by the user, so that the interactive interface can be customized according to the user's usage habits, and then generate a design drawing that satisfies the user, improve the user experience, and achieve the technical effect of customized interactive interface, thereby solving the technical problem that traditional interactive interfaces are highly complex and difficult to match user usage habits.

[0127] It should be noted that the acquisition unit 301, input unit 302, and reconstruction unit 303 correspond to steps S201 to S203 in Example 1. The three modules and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Example 1. It should be noted that the modules or units can be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The modules can also be part of a device and can be run in the computer terminal 10 provided in Example 1.

[0128] Example 3

[0129] An embodiment of the present application may provide an electronic device, Figure 4 This is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 Only one is shown) processor 402, memory 404, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0130] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0131] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: responding to the reconstruction instruction acting on the interactive interface, obtaining feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface; inputting the feature description information and the random noise vector into the target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training the deep learning model using interface design element samples and feature description information samples; responding to the selection operation for at least one design drawing, selecting a target design drawing from the at least one design drawing, and reconstructing the interactive interface based on the target design drawing.

[0132] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: in response to the adjustment instruction of the target design drawing, obtain the adjustment description information; adjust the target design drawing based on the adjustment description information to obtain the adjusted target design drawing; and update the target design drawing using the adjusted target design drawing.

[0133] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: adjusting the style characteristics of the interactive interface based on the style characteristics of the target design drawing; adjusting the layout structure of the interactive interface based on the layout structure of the target design drawing.

[0134] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: in response to the deep learning model being a generative adversarial network model, the generative adversarial network model is trained using interface design element samples and feature description information samples to obtain a target generative model.

[0135] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: initialize the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; use the interface design element samples to train the discriminator, wherein during the training of the discriminator, the model parameters of the generator remain fixed; in response to the completion of the discriminator training, use the interface design element samples and feature description information samples to train the generator, wherein during the training of the generator, the model parameters of the discriminator remain fixed.

[0136] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: input the interface design elements and interface design element samples randomly generated by the generator into the discriminator for identification to obtain a first identification result, wherein the first identification result is used to evaluate the ability of the discriminator to distinguish between randomly generated interface design elements and interface design element samples; in response to the identification result not meeting the identification performance index, adjust the model parameters of the discriminator, wherein the identification performance index is used to represent the correct classification rate of the discriminator for the interface design element samples and the incorrect classification rate of the randomly generated interface design elements; input the interface design elements and interface design element samples randomly generated by the generator into the adjusted discriminator for identification to obtain a second identification result of the discriminator; in response to the second identification result meeting the identification performance index, determine that the discriminator training is completed.

[0137] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: input the feature description information sample and the random noise vector into the generator to obtain a first predicted interface design element; input the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; in response to the first evaluation result not meeting the generation performance standard, adjust the model parameters of the generator to obtain an adjusted generator; input the feature description information sample and the random noise vector into the adjusted generator to obtain a second predicted interface design element; input the second predicted interface design element and the interface design element sample into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; in response to the second evaluation result meeting the generation performance index, determine that the generator training is completed.

[0138] An embodiment of the present application provides a method for adjusting an interactive interface. By responding to the user's reconstruction instructions and feature description information, it is possible to generate at least one design drawing that matches the user's preferences and needs, and then adjust the interactive interface according to the target design drawing selected by the user, so that the interactive interface can be customized according to the user's usage habits, thereby generating a design drawing that satisfies the user, improving the user experience, and achieving the technical effect of a customized interactive interface, thereby solving the technical problem that traditional interactive interfaces are highly complex and difficult to match user usage habits.

[0139] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, abbreviated as MID), a personal access device (Personal Access Device, abbreviated as PAD), etc. Figure 4 It does not limit the structure of the above electronic device. For example, the electronic device may also include Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 4 Different configurations shown.

[0140] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0141] Example 4

[0142] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the interactive interface adjustment method provided in the first embodiment.

[0143] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0144] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the method for adjusting the interactive interface.

[0145] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0146] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0147] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0148] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0149] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0150] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program code.

[0151] The above is only a preferred embodiment of the present application. 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 present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for adjusting an interactive interface, characterized in that: include: In response to a reconstruction instruction acting on an interactive interface, obtaining feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface; Inputting the feature description information and the random noise vector into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples; In response to a selection operation on the at least one design drawing, a target design drawing is selected from the at least one design drawing, and the interactive interface is reconstructed based on the target design drawing.

2. The method according to claim 1, characterized in that Before reconstructing the interactive interface based on the target design diagram, the method further includes: In response to an adjustment instruction for the target design drawing, obtaining adjustment description information; Adjust the target design drawing based on the adjustment description information to obtain the adjusted target design drawing; The target design diagram is updated using the adjusted target design diagram.

3. The method according to claim 1, characterized in that Reconstructing the interactive interface based on the target design diagram includes: Adjusting the style characteristics of the interactive interface based on the style characteristics of the target design; Based on the layout structure of the target design diagram, the layout structure of the interactive interface is adjusted.

4. The method according to claim 1, wherein The method further comprises: In response to the deep learning model being a generative adversarial network model, the generative adversarial network model is trained using the interface design element samples and the feature description information samples to obtain the target generation model.

5. The method according to claim 4, characterized in that The generative adversarial network model includes a generator and a discriminator. The generative adversarial network model is trained using the interface design element samples and the feature description information samples to obtain the target generation model, including: Initializing the model parameters of the generator and the model parameters of the discriminator, wherein the model parameters include at least weight parameters and bias parameters; Using the interface design element samples, training the discriminator, wherein during the training of the discriminator, the model parameters of the generator remain fixed; In response to the completion of the discriminator training, the generator is trained using the interface design element samples and the feature description information samples, wherein during the training of the generator, the model parameters of the discriminator remain fixed.

6. The method according to claim 5, characterized in that Using the interface design element sample, training the discriminator includes: Inputting the randomly generated interface design elements and the interface design element samples into the discriminator for recognition to obtain a first recognition result, wherein the first recognition result is used to evaluate the ability of the discriminator to distinguish between the randomly generated interface design elements and the interface design element samples; In response to the first recognition result failing to meet a recognition performance index, adjusting a model parameter of the discriminator, wherein the recognition performance index is used to represent a correct classification rate of the discriminator for the interface design element samples and an incorrect classification rate of the randomly generated interface design elements; Inputting the interface design elements randomly generated by the generator and the interface design element samples into the adjusted discriminator for recognition, thereby obtaining a second recognition result of the discriminator; In response to the second recognition result reaching the recognition performance index, it is determined that the discriminator training is completed.

7. The method according to claim 5, characterized in that Training the generator using the interface design element sample and the feature description information sample includes: Inputting the feature description information sample and the random noise vector into the generator to obtain a first prediction interface design element; Inputting the first predicted interface design element and the interface design element sample corresponding to the feature description information sample into the trained discriminator for evaluation to obtain a first evaluation result, wherein the first evaluation result is used to represent the similarity between the first predicted interface design element and the interface design element sample; In response to the first evaluation result failing to meet a generation performance standard, adjusting a model parameter of the generator to obtain an adjusted generator; Inputting the feature description information sample and the random noise vector into the adjusted generator to obtain a second prediction interface design element; Inputting the second predicted interface design element and the interface design element sample into the trained discriminator for evaluation to obtain a second evaluation result, wherein the second evaluation result is used to represent the similarity between the second predicted interface design element and the interface design element sample; In response to the second evaluation result reaching the generation performance index, it is determined that the generator training is completed.

8. A device for adjusting an interactive interface, characterized in that: include: an acquiring unit, configured to respond to a reconstruction instruction acting on an interactive interface and acquire feature description information of the interactive interface, wherein the feature description information is used to describe the interactive interface; an input unit, configured to input the feature description information and the random noise vector into a target generation model for analysis to obtain at least one design drawing corresponding to the feature description information, wherein the target generation model is obtained by pre-training a deep learning model using interface design element samples and feature description information samples; The reconstruction unit is configured to select a target design from the at least one design in response to a selection operation on the at least one design, and to reconstruct the interactive interface based on the target design.

9. An electronic device, characterized in that: include: a memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 7 when running.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The method comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images