Image data processing method, device, storage medium and computer equipment

Abstract

The present invention discloses a method, device, storage medium and computer equipment for processing image data. The method comprises: collecting shooting data of a shooting object, wherein the shooting data comprises at least one of the following: shooting pictures and shooting videos; analyzing the shooting data, determining a filter model matching the shooting data from a plurality of filter models, wherein the filter model is an adversarial model; using the selected filter model to perform filter processing on the shooting data to generate a filter image. The present invention solves the technical problem that the user end has high professional requirements for image processing and low processing efficiency.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to an image data processing method, device, storage medium and computer equipment. Background Art

[0002] In the field of image processing, after users take images, the imaging effects of the images and videos they take are not full and vivid enough due to many external factors, such as the lack of prominent contrast of the image subject, the lack of saturated color of the image, overexposure or underexposure of the imaging device, etc. In order to improve the imaging effect, users can use image processing software to process the image, or retake the image. However, among the above methods, the former requires high professional skills and is time-consuming and laborious, while the latter is inefficient and reduces user enthusiasm.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a method, apparatus, storage medium and computer equipment for processing image data, so as to at least solve the technical problems that the user end has high professional requirements for image processing and low processing efficiency.

[0005] According to one aspect of an embodiment of the present invention, there is provided a method for processing image data, comprising: collecting shooting data of a shooting object, wherein the shooting data comprises at least one of the following: shooting pictures and shooting videos; analyzing the shooting data, and determining a filter model that matches the shooting data from a plurality of types of filter models, wherein the filter model is an adversarial model; and performing filter processing on the shooting data using the selected filter model to generate a filtered image.

[0006] According to another aspect of an embodiment of the present invention, a method for processing image data is also provided, comprising: displaying shooting data collected by a shooting device on an interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; if a filter instruction is detected in any area of ​​the interactive interface, triggering analysis of the screen of the shooting data, and determining a filter model that matches the shooting data, wherein the filter model is an adversarial model; and displaying a filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

[0007] According to another aspect of an embodiment of the present invention, a method for processing image data is also provided, comprising: displaying shooting data on an interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; sensing a filter instruction matching the shooting data in the interactive interface; responding to the filter instruction, determining a filter model matching the shooting data, wherein the filter model is an adversarial model; outputting a selection page on the interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to characterize filter models of different levels used for the shooting data; and displaying a filter image on the interactive interface, wherein the filter image is an image obtained after filtering the shooting data based on the selected filter model.

[0008] According to another aspect of an embodiment of the present invention, there is also provided a method for processing image data, including: a front-end client uploading shooting data of a shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the front-end client transmitting the shooting data to a back-end server; the front-end client receiving a filter model returned by the back-end server that matches the shooting data, wherein the filter model is an adversarial model determined from a plurality of types of filter models; the front-end client uses the selected filter model to perform filter processing on the shooting data to generate a filtered image.

[0009] According to one aspect of an embodiment of the present invention, there is provided an image data processing device, comprising: a first acquisition module, used to acquire photographic data of a photographic object, wherein the photographic data comprises at least one of the following: photographed pictures and photographed videos; a first determination module, used to analyze the photographic data, and determine a filter model that matches the photographic data from a plurality of types of filter models, wherein the filter model is an adversarial model; and a first generation module, used to perform filter processing on the photographic data using the selected filter model to generate a filtered image.

[0010] According to another aspect of an embodiment of the present invention, there is also provided an image data processing device, comprising: a first display module, used to display shooting data collected by a shooting device on an interactive interface, wherein the shooting data includes at least one of the following: shot pictures and shot videos; a second determination module, used to trigger analysis of the picture of the shooting data when a filter instruction is detected in any area of ​​the interactive interface, and determine a filter model that matches the shooting data, wherein the filter model is an adversarial model; and a second display module, used to display a filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

[0011] According to another aspect of an embodiment of the present invention, there is also provided an image data processing device, comprising: a third display module, used to display shooting data on an interactive interface, wherein the shooting data includes at least one of the following: shot pictures and shot videos; a first sensing module, used to sense a filter instruction matching the shooting data in the interactive interface; a third determination module, used to respond to the filter instruction and determine a filter model matching the shooting data, wherein the filter model is an adversarial model; a first output module, used to output a selection page on the interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; and a fourth display module, used to display a filter image on the interactive interface, wherein the filter image is an image obtained after filtering the shooting data based on the selected filter model.

[0012] According to another aspect of an embodiment of the present invention, there is further provided an image data processing device, comprising: a first uploading module, used by a front-end client to upload shooting data of a shooting object, wherein the shooting data comprises at least one of the following: shooting pictures and shooting videos; a first transmission module, used by the front-end client to transmit the shooting data to a back-end server; a first receiving module, used by the front-end client to receive a filter model matched with the shooting data returned by the back-end server, wherein the filter model is an adversarial model determined from a plurality of types of filter models; and a second generating module, used by the front-end client to filter the shooting data using the selected filter model to generate a filtered image.

[0013] According to one aspect of an embodiment of the present invention, a storage medium is further provided, the storage medium comprising a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute any one of the above-mentioned methods for processing image data.

[0014] According to another aspect of an embodiment of the present invention, a computer device is also provided, including: a memory and a processor, the memory storing a computer program; the processor is used to execute the computer program stored in the memory, and when the computer program is run, the processor executes any one of the above-mentioned methods for processing image data.

[0015] In an embodiment of the present invention, a method of collecting shooting data of a shooting object is adopted, and a filter model matching the shooting data is determined and the shooting data is filtered using the filter model, so as to achieve the purpose of generating a filter image processed by the filter model, thereby achieving the technical effect of improving the imaging effect of the shot image, and further solving the technical problem of high professional requirements for image processing on the user side and low processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of a computer terminal for implementing an image data processing method according to an embodiment of the present invention;

[0018] Figure 2 is a flowchart of a first method for processing image data provided by an embodiment of the present invention;

[0019] Figure 3 is a flowchart of a second method for processing image data provided by an embodiment of the present invention;

[0020] Figure 4 is a flowchart of a third method for processing image data provided by an embodiment of the present invention;

[0021] Figure 5 is a flowchart of a fourth method for processing image data provided by an embodiment of the present invention;

[0022] Figure 6 is a schematic diagram of a method for processing image data provided according to an optional embodiment of the present invention;

[0023] Figure 7 is a structural block diagram of a first image data processing device provided according to an embodiment of the present invention;

[0024] Figure 8 is a structural block diagram of a second image data processing device provided according to an embodiment of the present invention;

[0025] Fig. 9 is a structural block diagram of a third image data processing device provided according to an embodiment of the present invention;

[0026] Fig.10 is a structural block diagram of an image data processing device 4 provided according to an embodiment of the present invention;

[0027] Fig.11 It is a structural block diagram of a computer terminal according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below 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 of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

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

[0030] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:

[0031] Filter is a method of processing image data. It processes the pixel values ​​in the image to achieve various special effects of the image.

[0032] Smart filters automatically add suitable filter effects to an image through machine learning, without the need for manual selection of filters and adjustment of their effects. Usually, it is necessary to jointly process image elements such as image channels, pixels, and layers, so as to strengthen some parts of the image / video and weaken some parts, thereby obtaining visual effects such as gradients, halos, and tones, so that the entire image / video meets the requirements of the human aesthetic perception system and ultimately achieves better artistic effects.

[0033] U-net, an algorithm for semantic segmentation using a fully convolutional network, refers to the paper<ConvolutionalNetworks for Biomedical Image Segmentation> A network structure proposed in .

[0034] Generative Adversarial Network (GAN) is an unsupervised learning algorithm that learns by letting two neural networks compete with each other to obtain a neural network model (i.e., adversarial model) that meets the requirements.

[0035] Batch normalization layer, a theoretical technique for improving the performance and training stability of deep neural networks, overcoming the problem of slow convergence as the number of neural network layers increases. This technique can present zero mean / unit variance inputs to any layer in the neural network.

[0036] Example 1

[0037] According to an embodiment of the present invention, a method embodiment of processing image data 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.

[0038] 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 1 FIG. 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for processing image data. Figure 1 As shown, the computer terminal 10 (or mobile device 10) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (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 can 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 as shown, or with Figure 1 Different configurations are shown.

[0039] 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 circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0040] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the image data processing method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the image data processing method described above is realized. 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 examples, the memory 104 may further include a memory remotely arranged 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.

[0041] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0042] Under the above operating environment, this application provides Figure 2 The method for processing the image data shown. Figure 2 is a flowchart of a first method for processing image data according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0043] Step S202, collecting shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos;

[0044] Step S204, analyzing the shooting data, and determining a filter model that matches the shooting data from a plurality of types of filter models, wherein the filter model is an adversarial model;

[0045] Step S206: Use the selected filter model to perform filter processing on the captured data to generate a filtered image.

[0046] Through the above-mentioned processing, the shooting data of the shooting object is collected, and the filter model matching the shooting data is determined and the shooting data is filtered using the filter model, so as to achieve the purpose of generating a filter image processed by the filter model, thereby achieving the technical effect of improving the imaging effect of the shot image, and further solving the technical problem of high professional requirements for image processing on the user side and low processing efficiency.

[0047] As an optional embodiment, determining the filter model that matches the shooting data includes: classifying the shooting data into scenes to obtain the scene category to which the shooting data belongs; and based on the scene category to which the shooting object belongs in the shooting data, calling the filter model that matches the shooting data from a plurality of filter models. The scene category to which the shooting data belongs is important for the selection of the filter model, because the shooting objects in different scenes have different requirements for image style, overall color tone, light and other factors, and therefore cannot apply a unified image processing model.

[0048] Optionally, the present embodiment may first classify the scene of the captured data and obtain the scene category. For example, the scene category may include: food, clothing, outdoor, digital products, etc. The recognition and classification of scenes may include a variety of methods. For example, an algorithm model that can recognize and classify image scenes may be pre-trained through artificial intelligence learning methods, and the image data may be put into the algorithm model. The scene category to which the captured image belongs may be obtained through the processing of the algorithm model; the features in the image may also be extracted through the processing of the image data through automated classification. When the extracted image features match the feature data in a pre-set scene feature library, the scene category to which the captured image belongs is determined.

[0049] This embodiment can also call a filter model that matches the scene category. For example, in the "food" scene, call a food filter model that matches the food image. The food filter model can improve the contrast and saturation of the foreground, so that the food image in the image after model processing is more colorful and full, the details are clearer, the food theme is more prominent, and the imaging effect is better. For another example, in the "outdoor" scene, call an outdoor filter model that matches the outdoor environment. The outdoor filter model can enhance the blur of the distant view and background in the image, avoid too many interference factors in the outdoor environment, and enhance the presence of the subject in the picture.

[0050] As an optional embodiment, the scene category to which the shooting data belongs can be obtained by the following method: extracting image features from the shooting data, wherein the image features include at least one of the following: features of the shooting object and features of the background image; determining the scene parameters of the shooting data based on the image features, wherein the scene parameters are used to characterize the product category to which the shooting object recorded in the shooting data belongs; and constructing the scene category to which the shooting data belongs based on the scene parameters of the shooting data. Through the above processing, by extracting the object image features and background image features of the shooting data, both the type of the shooting object and the type of the environment in which the shooting object is located can be considered, thereby achieving the technical effect of accurately determining the scene category of the shooting data. The features of the shooting object can include: the category, color, size, and shape of the object. For example, the category of the object can be cake, the color is white, the size is 50% of the image area, and the shape is round; the features of the background image can include: light, brightness, background scene, background objects included, etc. For example, the background can be clear light, moderate brightness, the background scene is an indoor environment, and the background objects included are tables, chairs, etc. By extracting the above features, the scene parameters of the shooting data are determined, and the scene category is constructed according to the scene parameters. For example, the above scene category may be: indoor close-up food photography with sufficient light. By accurately determining the scene category of the photographed object, the subsequent filter model can process the image data more specifically and achieve better filter effects.

[0051] As an optional embodiment, the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types.

[0052] Optionally, the captured data is filtered using the selected filter model, including: extracting global elements from the captured data based on a U-Net model structure with global features in the filter model, wherein the global elements include at least one of the following: light, composition, foreground and background; according to the global elements, migrating the pixel data distribution of the captured data to the pixel data distribution of a scene type matching the captured data; and generating a filtered image based on the migration result.

[0053] Each filter model can process image data matching the scene type for different scene types. The U-Net model structure with global features can extract global elements of light, composition, foreground and background light from the shooting data, realizing more accurate and uniform image data transformation. In the related art, the generative adversarial network can extract local features of the image through the local feature U-Net network structure, and use the local features to process the image at the pixel level, thereby realizing the filter processing of the image. However, due to the lack of extraction and processing of global features of the image, the global elements such as the overall light, composition, foreground, and background of the image are not considered, resulting in the situation that the image area that needs to be strengthened is not strengthened, and the image area that needs to be weakened is strengthened, so that the generated filter image is difficult to obtain a satisfactory artistic effect. Through the U-Net model structure with global features, the features that reflect the scene category of the shot image can be extracted more accurately and comprehensively, and the pixel data of the shot data can be processed in a manner that conforms to the characteristics of the scene category. Specifically, the pixel data distribution of the captured data is migrated to the pixel data distribution of the scene type that matches the captured data, thereby enhancing the integrity of the image data processed by the filter model, making the processing of the filter model more consistent with the style of the scene category, and avoiding inconsistency in image style. The independent batch normalization layer can normalize the sample features in the training data set, making the gradient direction of the model more accurate and effective when converging, thereby overcoming the problem of slower convergence speed as the number of neural network layers increases, allowing the model to perform gradient descent more boldly, making it easier to obtain a filter model through training, and the model itself is more stable. For example, after the network structure of the filter model includes an independent batch normalization layer, the filter models for different scene categories will be trained faster, resulting in a filter model with better classification effect.

[0054] As an optional embodiment, after determining the filter model matching the shooting data from multiple types of filter models, the method includes: outputting a selection page on an interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to characterize filter models of different levels used for the shooting data; and obtaining a filter model matching the filter option when any filter option is triggered. After determining the filter model matching the shooting data, the user can be allowed to independently select the degree of processing of the image data by the determined filter model, thereby enhancing the user's autonomous selectivity and facilitating the user to obtain an image that better meets their needs. For example, when the filter model has been determined, the interactive interface may provide multiple options such as "preliminary filter processing", "moderate filter processing" and "high filter processing", and corresponding to different filter options, the filter model may adjust the parameter settings within the model to process the original image data to different degrees. When any filter option is triggered, the corresponding filter model automatically changes the internal parameters according to the triggered filter option, and completes the call to the adjusted filter model.

[0055] Figure 3 FIG. 2 is a flowchart of a second method for processing image data according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0056] Step S302, displaying shooting data collected by the shooting device on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos;

[0057] Step S304, if a filter instruction is detected in any area of ​​the interactive interface, triggering a picture of the shooting data to analyze and determine a filter model that matches the shooting data, wherein the filter model is an adversarial model;

[0058] Step S306: displaying a filter image on the interactive interface, wherein the filter image is an image generated by performing filter processing on the captured data using the selected filter model.

[0059] Through the above-mentioned processing, the shooting data is displayed on the interactive interface, and the pictures and filter images of the shooting data are analyzed. By determining the filter model according to the detected filter instructions and using the filter model to process the shooting data, the process of processing the image data using the filter model is displayed on the interactive interface, achieving the technical effect of intuitively displaying the filter effect, and bringing a better user experience to the user.

[0060] Figure 4 1 is a flowchart of a third method for processing image data according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0061] Step S402, displaying shooting data on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos;

[0062] Step S404, a filter instruction matching the shooting data is sensed in the interactive interface;

[0063] Step S406, responding to the filter instruction, determining a filter model that matches the captured data, wherein the filter model is an adversarial model;

[0064] Step S408, outputting a selection page on the interactive interface, the selection page providing at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data;

[0065] Step S410: displaying a filter image on the interactive interface, wherein the filter image is an image obtained by filtering the captured data based on the selected filter model.

[0066] Through the above processing, it is possible to determine the specific filter model on the interactive interface based on the filter instructions corresponding to the filter model type and the filter options used to characterize filter models of different levels, and filter processing is performed on the shooting data according to the determined filter model, thereby achieving the technical effect of providing an interactive means for selecting the type of filter model and the level of the filter model.

[0067] As an optional embodiment, if the picture quality of the captured data is lower than the standard data, a filter instruction matching the captured data is triggered. When the picture quality of the captured data is too low, the conventional filter model may not be able to effectively process the captured data. At this time, the above processing can trigger a filter instruction to call a filter model matching the picture quality of the captured data with the current picture quality. For example, the called filter model can be a model that can process the filter data according to the picture quality of the current captured data, or it can be a model that can improve the picture quality of the current captured data according to a predetermined algorithm and then process the filter data.

[0068] As an optional embodiment, before displaying the filter image on the interactive interface, the method further includes: receiving a selection instruction in the area where the corresponding filter option is located on the selection page; responding to the selection instruction, triggering the corresponding filter option, and calling the filter model of the corresponding level. Through the above processing, an interactive means is provided for the user, so that the user can independently select the level of the filter model and obtain a filter image that meets the needs.

[0069] Figure 5 FIG. 4 is a flowchart of a fourth method for processing image data according to an embodiment of the present invention. Figure 5 As shown, the process includes the following steps:

[0070] Step S502, the front-end client uploads the shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos;

[0071] Step S504, the front-end client transmits the captured data to the back-end server;

[0072] Step S506, the front-end client receives the filter model matched with the shooting data returned by the back-end server, wherein the filter model is an adversarial model determined from a plurality of types of filter models;

[0073] Step S508: The front-end client uses the selected filter model to perform filter processing on the captured data to generate a filtered image.

[0074] Through the above processing, the front-end client can call the model from the back-end server. The back-end server returns a filter model that matches the shooting data transmitted by the front-end client. The front-end client uses the filter model to process the shooting data and generate a filtered image.

[0075] Figure 6 FIG. 1 is a schematic diagram of a method for processing image data according to an optional embodiment of the present invention. Figure 6 As shown, the image processing method according to an optional implementation manner of the present invention includes two major units: front-end interaction and background algorithm.

[0076] Figure 6 The process in the front-end interaction unit includes the following steps:

[0077] S1, merchants upload images or videos; S2, scene / category information display, by using the scene classification module to analyze the scene / category of the image and display the scene category corresponding to the image; S3, smart filter processing, by using the filter model corresponding to the scene category to which the image belongs in the smart filter module to process the image data; S4, obtain the smart filter image.

[0078] Figure 6 The backend algorithm unit in the app includes a scene classification module and an intelligent filter module. The scene classification module predicts the scene category to which the image belongs by analyzing the image uploaded by the user, and uses different filter models to process the image data according to the obtained scene category. Among them, the filter model is an adversarial model, including a generative adversarial network, which includes a U-Net network structure of global features and an independent batch normalization layer for a specific style.

[0079] Generative adversarial network is a neural network learning algorithm that can use the original image as the input sample image to generate a new image with a filter effect. Generative adversarial network consists of two parts: generator and discriminator. The generator generates a dataset of images processed by filters based on the original image data as a sample set through an encoder-decoder; the discriminator is used to compare the images before and after processing to distinguish whether the image is the original real image or the image generated by the generator. The generator and the discriminator can continuously improve their respective performance in the above confrontation, so as to achieve a dynamic balance and obtain an adversarial model that meets the requirements. Generative adversarial network can use max-min loss function to constrain the learning process of the model. By adding a U-Net network structure based on global features and an independent batch normalization layer with a specific style to the generative adversarial network, the generative adversarial network can be optimized to guide the adversarial model to learn features with global and specific styles. For example, the generative adversarial network with food as the theme is trained to obtain an adversarial model with food as the theme. The model can perform targeted processing on images with food as the scene category, so that the overall style of the processed image remains consistent and is very consistent with the food theme.

[0080] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that the image data processing method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0082] Example 2

[0083] According to an embodiment of the present invention, a device for implementing the above-mentioned image data processing method 1 is also provided. Figure 7 is a structural block diagram of an image data processing device according to an embodiment of the present invention. Figure 7As shown, the device includes: a first acquisition module 702, a first determination module 704 and a first generation module 706. The device is described below.

[0084] The first acquisition module 702 is used to acquire shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the first determination module 704 is connected to the above-mentioned first acquisition module 702, and is used to analyze the shooting data and determine a filter model that matches the shooting data from a plurality of types of filter models, wherein the filter model is an adversarial model; the first generation module 706 is connected to the above-mentioned first determination module 704, and is used to use the selected filter model to filter the shooting data to generate a filter image.

[0085] It should be noted that the first acquisition module 702, the first determination module 704 and the first generation module 706 correspond to steps S202 to S206 in Example 1, and the three modules and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules as part of the device can be run in the computer terminal 10 provided in Example 1.

[0086] According to an embodiment of the present invention, a device for implementing the second method for processing image data is also provided. Figure 8 is a structural block diagram of a second apparatus for processing image data according to an embodiment of the present invention. Figure 8 As shown, the device includes: a first display module 802, a second determination module 804 and a second display module 806. The device is described below.

[0087] The first display module 802 is used to display the shooting data collected by the shooting device on the interactive interface, wherein the shooting data includes at least one of the following: shot pictures and shot videos; the second determination module 804 is connected to the above-mentioned first display module 802, and is used to trigger the analysis of the screen of the shooting data when a filter instruction is detected in any area of ​​the interactive interface, and determine the filter model that matches the shooting data, wherein the filter model is an adversarial model; the second display module 806 is connected to the above-mentioned second determination module 804, and is used to display the filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

[0088] It should be noted that the first display module 802, the second determination module 804 and the second display module 806 correspond to steps S302 to S306 in Example 1, and the three modules and the 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 above modules, as part of the device, can be run in the computer terminal 10 provided in Example 1.

[0089] According to an embodiment of the present invention, a device for implementing the third method for processing image data is also provided. Fig. 9 is a structural block diagram of a third apparatus for processing image data according to an embodiment of the present invention. Fig. 9 As shown, the device includes: a third display module 902, a first sensing module 904, a third determination module 906, a first output module 908 and a fourth display module 910. The device is described below.

[0090] The third display module 902 is used to display the shooting data on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the first sensing module 904 is connected to the third display module 902, and is used to sense the filter instructions matching the shooting data in the interactive interface; the third determination module 906 is connected to the first sensing module 904, and is used to display the filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model; the first output module 908 is connected to the third determination module 906, and is used to output a selection page on the interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; the fourth display module 910 is connected to the first output module 908, and is used to display the filter image on the interactive interface, wherein the filter image is an image obtained by filtering the shooting data based on the selected filter model.

[0091] It should be noted that the third display module 902, the first sensing module 904, the third determination module 906, the first output module 908 and the fourth display module 910 correspond to steps S402 to S410 in Example 1, and the five modules and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules as part of the device can be run in the computer terminal 10 provided in Example 1.

[0092] According to an embodiment of the present invention, a device for implementing the fourth method for processing image data is also provided. Fig.10 is a structural block diagram of an image data processing device 4 according to an embodiment of the present invention. Fig.10As shown, the device includes: a first uploading module 1002, a first transmission module 1004, a first receiving module 1006 and a second generating module 1008. The device is described below.

[0093] The first uploading module 1002 is used for the front-end client to upload the shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the first transmission module 1004 is connected to the above-mentioned first uploading module 1002, and is used for the front-end client to transmit the shooting data to the back-end server; the first receiving module 1006 is connected to the above-mentioned first transmission module 1004, and is used for the front-end client to receive the filter model matching the shooting data returned by the back-end server, wherein the filter model is an adversarial model determined from multiple types of filter models; the second generating module 1008 is connected to the above-mentioned first receiving module 1006, and is used for the front-end client to use the selected filter model to filter the shooting data to generate a filter image.

[0094] It should be noted that the first uploading module 1002, the first transmission module 1004, the first receiving module 1006 and the second generating module 1008 correspond to steps S502 to S508 in Example 1, and the examples and application scenarios implemented by the four modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules as part of the device can be run in the computer terminal 10 provided in Example 1.

[0095] Example 3

[0096] The embodiment of the present invention can provide a computer terminal, which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal can also be replaced by a terminal device such as a mobile terminal.

[0097] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of the computer network.

[0098] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the image data processing method of the application: collecting shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; analyzing the shooting data, and determining a filter model that matches the shooting data from multiple types of filter models, wherein the filter model is an adversarial model; using the selected filter model to filter the shooting data to generate a filter image.

[0099] Optionally, Fig.11 is a structural block diagram of a computer terminal according to an embodiment of the present invention. Fig.11As shown, the computer terminal may include: one or more (only one is shown in the figure) processors 1102, a memory 1104, etc.

[0100] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the image data processing method and device in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned image data processing 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 combinations thereof.

[0101] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: collect shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; analyze the shooting data, and determine a filter model that matches the shooting data from multiple types of filter models, wherein the filter model is an adversarial model; use the selected filter model to filter the shooting data to generate a filter image.

[0102] Optionally, the processor may also execute program code for the following steps: analyzing the shooting data, and determining a filter model that matches the shooting data from a plurality of types of filter models, including: performing scene classification on the shooting data, and obtaining the scene category to which the shooting data belongs; and based on the scene category to which the shooting object belongs in the shooting data, calling a filter model that matches the shooting data from a plurality of types of filter models.

[0103] Optionally, the processor may also execute program code for the following steps: performing scene classification on the captured data to obtain the scene category to which the captured data belongs, including: extracting image features in the captured data, wherein the image features include at least one of the following: features of the captured object and features of the background image; based on the image features, determining scene parameters of the captured data, wherein the scene parameters are used to characterize the product category to which the captured object recorded in the captured data belongs; and based on the scene parameters of the captured data, constructing the scene category to which the captured data belongs.

[0104] Optionally, the processor may also execute program code of the following steps: the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types.

[0105] Optionally, the processor may also execute program code of the following steps: performing filter processing on the captured data using a selected filter model, including: extracting global elements from the captured data based on a U-Net model structure with global features in the filter model, wherein the global elements include at least one of the following: light, composition, foreground, and background; migrating the pixel data distribution of the captured data to the pixel data distribution of a scene type matching the captured data based on the global elements; and generating a filtered image based on the migration result.

[0106] Optionally, the processor may also execute program code for the following steps: after determining a filter model that matches the shooting data from a plurality of types of filter models, the process includes: outputting a selection page on an interactive interface, the selection page providing at least one filter option, wherein different filter options are used to characterize filter models of different levels used for the shooting data; and obtaining a filter model that matches the filter option when any filter option is triggered.

[0107] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: displaying the shooting data collected by the shooting device on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; if a filter instruction is detected in any area of ​​the interactive interface, triggering the picture of analyzing the shooting data, and determining the filter model that matches the shooting data, wherein the filter model is an adversarial model; displaying the filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

[0108] The processor can call the information and application programs stored in the memory through the transmission device to execute the following steps: displaying the shooting data on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; sensing the filter instructions matching the shooting data in the interactive interface; responding to the filter instructions, determining the filter model matching the shooting data, wherein the filter model is an adversarial model; outputting a selection page on the interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; and displaying the filter image on the interactive interface, wherein the filter image is an image obtained after filtering the shooting data based on the selected filter model.

[0109] Optionally, the processor may further execute program code of the following steps: if the image quality of the captured data is lower than that of the standard data, trigger a filter instruction matching the captured data.

[0110] Optionally, the processor may also execute the program code of the following steps: before displaying the filter image on the interactive interface, the method also includes: receiving a selection instruction in the area where the corresponding filter option is located on the selection page; responding to the selection instruction, triggering the corresponding filter option, and calling the filter model of the corresponding level.

[0111] The processor can call the information and application stored in the memory through the transmission device to execute the following steps: the front-end client uploads the shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the front-end client transmits the shooting data to the background server; the front-end client receives the filter model matched with the shooting data returned by the background server, wherein the filter model is an adversarial model determined from multiple types of filter models; the front-end client uses the selected filter model to filter the shooting data to generate a filter image.

[0112] In the embodiment of the present invention, the shooting data of the shooting object is collected, and the filter model matching the shooting data is determined and the shooting data is filtered using the filter model, so as to achieve the purpose of generating a filter image processed by the filter model, thereby achieving the technical effect of improving the imaging effect of the shot image, and further solving the technical problem of high professional requirements for image processing on the user side and low processing efficiency.

[0113] It can be understood by those skilled in the art that Fig.11 The structure shown is for illustration only, and the computer terminal may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Fig.11 It does not limit the structure of the above electronic device. For example, the computer terminal may also include Fig.11 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Fig.11 Different configurations are shown.

[0114] A person of ordinary skill in the art can 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, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0115] Example 4

[0116] The embodiment of the present invention further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the image data processing method provided in the first embodiment.

[0117] Optionally, in this embodiment, the above 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.

[0118] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: acquiring shooting data of a shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; analyzing the shooting data, and determining a filter model that matches the shooting data from a plurality of types of filter models, wherein the filter model is an adversarial model; and using the selected filter model to filter the shooting data to generate a filter image.

[0119] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: analyzing the shooting data, and determining a filter model that matches the shooting data from a plurality of types of filter models, including: performing scene classification on the shooting data, and obtaining the scene category to which the shooting data belongs; and based on the scene category to which the shooting object belongs in the shooting data, calling a filter model that matches the shooting data from a plurality of types of filter models.

[0120] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: performing scene classification on the shooting data to obtain the scene category to which the shooting data belongs, including: extracting image features in the shooting data, wherein the image features include at least one of the following: features of the shot object and features of the background image; based on the image features, determining scene parameters of the shooting data, wherein the scene parameters are used to characterize the product category to which the shot object recorded in the shooting data belongs; based on the scene parameters of the shooting data, constructing the scene category to which the shooting data belongs.

[0121] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types.

[0122] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: filtering the captured data using a selected filter model, including: extracting global elements from the captured data based on a U-Net model structure with global features in the filter model, wherein the global elements include at least one of the following: light, composition, foreground, and background; based on the global elements, migrating the pixel data distribution of the captured data to the pixel data distribution of a scene type that matches the captured data; and generating a filtered image based on the migration result.

[0123] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: after determining a filter model that matches the shooting data from a plurality of types of filter models, including: outputting a selection page on an interactive interface, the selection page providing at least one filter option, wherein different filter options are used to characterize filter models of different levels used for the shooting data; when any filter option is triggered, obtaining a filter model that matches the filter option.

[0124] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: displaying shooting data collected by the shooting device on an interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; if a filter instruction is detected in any area of ​​the interactive interface, triggering a screen for analyzing the shooting data, and determining a filter model that matches the shooting data, wherein the filter model is an adversarial model; displaying a filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

[0125] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: displaying shooting data on an interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; sensing a filter instruction that matches the shooting data in the interactive interface; responding to the filter instruction, determining a filter model that matches the shooting data, wherein the filter model is an adversarial model; outputting a selection page on the interactive interface, the selection page providing at least one filter option, wherein different filter options are used to characterize filter models of different levels used for the shooting data; displaying a filter image on the interactive interface, wherein the filter image is an image obtained by filtering the shooting data based on the selected filter model.

[0126] Optionally, in this embodiment, the storage medium is configured to store a program code for executing the following steps: if the image quality of the captured data is lower than the standard data, trigger a filter instruction matching the captured data.

[0127] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: before displaying the filter image on the interactive interface, the method also includes: receiving a selection instruction in the area where the corresponding filter option is located on the selection page; responding to the selection instruction, triggering the corresponding filter option, and calling the filter model of the corresponding level.

[0128] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: the front-end client uploads shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; the front-end client transmits the shooting data to the back-end server; the front-end client receives a filter model returned by the back-end server that matches the shooting data, wherein the filter model is an adversarial model determined from multiple types of filter models; the front-end client uses the selected filter model to filter the shooting data to generate a filter image.

[0129] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0130] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0131] 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. There may be other division methods in actual implementation. For example, 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.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0133] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0134] 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 invention, in essence, 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 a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0135] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for processing image data, characterized in that: include: Collecting shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; Analyze the captured data, and determine a filter model that matches the captured data from a plurality of filter models, wherein the filter model is an adversarial model, and the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; The shooting data is filtered using the selected filter model to generate a filtered image.

2. The method according to claim 1, characterized in that: Analyzing the shooting data and determining a filter model matching the shooting data from a plurality of types of filter models, including: Performing scene classification on the shooting data to obtain the scene category to which the shooting data belongs; Based on the scene category to which the photographed object belongs in the photographed data, a filter model matching the photographed data is called from a plurality of types of filter models.

3. The method according to claim 2, characterized in that Performing scene classification on the shooting data to obtain the scene category to which the shooting data belongs includes: Extracting image features from the photographed data, wherein the image features include at least one of the following: features of the photographed object and features of the background image; Based on the image features, determining a scene parameter of the shooting data, wherein the scene parameter is used to characterize a product category to which the shooting object recorded in the shooting data belongs; Based on the scene parameters of the shooting data, a scene category to which the shooting data belongs is constructed.

4. The method according to claim 1, characterized in that: Using the selected filter model to perform filter processing on the captured data includes: Based on the U-Net model structure with global features in the filter model, extracting global elements from the captured data, wherein the global elements include at least one of the following: light, composition, foreground and background; Migrating the pixel data distribution of the shooting data to the pixel data distribution of the scene type matching the shooting data according to the global element; Based on the migration result, the filter image is generated.

5. The method according to claim 1, characterized in that After determining a filter model matching the shooting data from a plurality of types of filter models, the method further comprises: Outputting a selection page on an interactive interface, the selection page providing at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; When any filter option is triggered, a filter model matching the filter option is obtained.

6. A method for processing image data, characterized in that: include: Displaying shooting data collected by the shooting device on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; If a filter instruction is detected in any area of ​​the interactive interface, a picture of the captured data is analyzed, and a filter model matching the captured data is determined, wherein the filter model is an adversarial model, and the network structure of the filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; A filter image is displayed on the interactive interface, wherein the filter image is an image generated by performing filter processing on the shooting data using the selected filter model.

7. A method for processing image data, characterized in that: include: Displaying shooting data on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; A filter instruction matching the shooting data is sensed in the interactive interface; In response to the filter instruction, determine a filter model that matches the captured data, wherein the filter model is an adversarial model, and the network structure of the filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; Outputting a selection page on the interactive interface, the selection page providing at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; A filter image is displayed on the interactive interface, wherein the filter image is an image obtained by filtering the captured data based on the selected filter model.

8. The method according to claim 7, characterized in that If the image quality of the captured data is lower than the standard data, a filter instruction matching the captured data is triggered.

9. The method according to claim 7, characterized in that: Before displaying the filtered image on the interactive interface, the method further includes: Receiving a selection instruction in the area where the corresponding filter option is located on the selection page; In response to the selection instruction, the corresponding filter option is triggered and the filter model of the corresponding level is called.

10. A method for processing image data, characterized in that: include: The front-end client uploads the shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; The front-end client transmits the captured data to the back-end server; The front-end client receives the filter model matched with the shooting data returned by the back-end server, wherein the filter model is an adversarial model determined from a plurality of types of filter models, and the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; The front-end client uses the selected filter model to perform filter processing on the shooting data to generate a filtered image.

11. An image data processing device, characterized in that: include: A first acquisition module, used to acquire shooting data of a shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; A first determination module is used to analyze the shooting data and determine a filter model matching the shooting data from a plurality of filter models, wherein the filter model is an adversarial model, and the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; The first generating module is used to use the selected filter model to perform filter processing on the shooting data to generate a filtered image.

12. An image data processing device, characterized in that: include: A first display module, used to display shooting data collected by the shooting device on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; A second determination module is used to trigger analysis of the picture of the shooting data and determine a filter model matching the shooting data when a filter instruction is detected in any area of ​​the interactive interface, wherein the filter model is an adversarial model, and the network structure of the filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; The second display module is used to display a filter image on the interactive interface, wherein the filter image is an image generated by filtering the shooting data using the selected filter model.

13. An image data processing device, characterized in that: include: A third display module, used to display shooting data on the interactive interface, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; A first sensing module, used for sensing a filter instruction matching the shooting data in the interactive interface; A third determination module is used to respond to the filter instruction and determine a filter model that matches the captured data, wherein the filter model is an adversarial model, and the network structure of the filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; A first output module, configured to output a selection page on the interactive interface, wherein the selection page provides at least one filter option, wherein different filter options are used to represent filter models of different levels used for the shooting data; A fourth display module is used to display a filter image on the interactive interface, wherein the filter image is an image obtained by filtering the captured data based on the selected filter model.

14. An image data processing device, characterized in that: include: A first uploading module, used for the front-end client to upload the shooting data of the shooting object, wherein the shooting data includes at least one of the following: shooting pictures and shooting videos; A first transmission module, used for the front-end client to transmit the captured data to a back-end server; A first receiving module is used for the front-end client to receive the filter model matched with the shooting data and returned by the back-end server, wherein the filter model is an adversarial model determined from a plurality of types of filter models, and the network structure of each filter model includes: a U-Net model structure of global features and independent batch normalization layers corresponding to different scene types; The second generating module is used for the front-end client to use the selected filter model to perform filter processing on the shooting data to generate a filtered image.

15. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the method for processing image data according to any one of claims 1 to 10.

16. A computer device, characterized in that: include: Memory and processor, The memory stores a computer program; The processor is used to execute the computer program stored in the memory, and when the computer program is run, the processor executes the image data processing method according to any one of claims 1 to 10.

Images