Toning processing method, storage medium and electronic equipment

Abstract

The invention discloses a toning processing method, a storage medium and electronic equipment. The method comprises the steps that a toning object is determined, and the toning object is an image to be toned or a video to be toned; obtaining a plurality of prediction masks and a plurality of prediction toning parameters corresponding to a plurality of toning areas in the toning object; performing toning processing on the plurality of prediction masks by adopting the plurality of prediction toning parameters to obtain a plurality of initial toning results; and performing fusion processing on the plurality of initial toning results to obtain a target toning result. The technical problems of low toning efficiency and poor toning effect caused by a complex operation process when toning is performed on an image and a video in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a color adjustment processing method, a storage medium, and an electronic device. Background Art

[0002] With the rapid development of the Internet and the popularity of smart phones, people can shoot and share images and videos more conveniently. However, due to the limitations of shooting conditions and environment, the images and videos shot often have color problems, such as saturation, contrast, unnatural hue, etc., which makes the presentation of images and videos unsatisfactory. Therefore, image and video color grading technology came into being, which can adjust the color of images and videos, improve the presentation effect, make them more vivid, beautiful and suitable for applications in specific scenarios. In the related technology, some professional editing tools have provided color grading functions, but manual adjustment is required, which is complicated and costly to use, and is not user-friendly for the general public. At the same time, the current preset color grading filters and end-to-end global color grading artificial intelligence (AI) algorithms cannot meet the professional colorist's needs for fine-grained and customized color grading. Therefore, how to lower the threshold for the use of professional editing software, so as to quickly complete the difficult color grading tasks and further reduce the color grading costs is of great research significance.

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

[0004] The embodiments of the present application provide a color adjustment processing method, a storage medium and an electronic device to at least solve the technical problems in the related art of low color adjustment efficiency and poor color adjustment effect caused by the complicated operation process when adjusting the color of images and videos.

[0005] According to one aspect of an embodiment of the present application, a color adjustment method is provided, including: determining a color adjustment object, wherein the color adjustment object is an image to be adjusted or a video to be adjusted; obtaining multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object; using multiple prediction color adjustment parameters to perform color adjustment on multiple prediction masks respectively to obtain multiple initial color adjustment results; and performing fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result.

[0006] According to another aspect of an embodiment of the present application, a color processing method is also provided, including: receiving a color object from a client, wherein the color object is an image to be colored or a video to be colored; obtaining multiple prediction masks and multiple prediction color parameters corresponding to multiple color areas in the color object, using multiple prediction color parameters to perform color processing on the multiple prediction masks respectively to obtain multiple initial color results, and fusing the multiple initial color results to obtain a target color result; and feeding back the target color result to the client.

[0007] According to another aspect of an embodiment of the present application, a computer-readable storage medium is further provided, the computer-readable storage medium including a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the color adjustment processing methods in the embodiments of the present application.

[0008] According to another aspect of an embodiment of the present application, an electronic device is also provided, including: a processor; and a memory, connected to the processor, for providing the processor with instructions for processing the following processing steps: determining a color adjustment object, wherein the color adjustment object is an image to be colored or a video to be colored; obtaining multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object; using multiple prediction color adjustment parameters to perform color adjustment on multiple prediction masks respectively to obtain multiple initial color adjustment results; and performing fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result.

[0009] In an embodiment of the present application, an image to be colored or a video to be colored is determined, and then multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas in the image to be colored or the video to be colored are obtained, and then multiple prediction coloring parameters are used to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results, and finally the multiple initial coloring results are fused to obtain a target coloring result, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the coloring efficiency of images and videos, improving the coloring effect, and reducing the coloring cost, thereby solving the technical problems of low coloring efficiency and poor coloring effect caused by the complicated operation process when coloring images and videos in the related art.

[0010] It is easy to notice that the above general description and the following detailed description are only for the purpose of exemplifying and explaining the present application, and do not constitute a limitation of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0012] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a color adjustment processing method is shown;

[0013] Figure 2 is a flow chart of a color adjustment method according to Embodiment 1 of the present application;

[0014] Figure 3 is a schematic diagram of a color adjustment method according to Example 1 of the present application;

[0015] Figure 4 is a schematic diagram of another color adjustment method according to Example 1 of the present application;

[0016] Figure 5 is a schematic diagram of a graphical user interface of a color adjustment processing method according to Example 1 of the present application;

[0017] Figure 6 is a flow chart of a color adjustment method according to Embodiment 2 of the present application;

[0018] Figure 7 is a schematic diagram of a color adjustment method on a cloud server according to Embodiment 2 of the present application;

[0019] Figure 8 is a structural block diagram of a color adjustment processing device according to Embodiment 1 of the present application;

[0020] Fig. 9 is a structural block diagram of a color adjustment processing device according to Embodiment 2 of the present application;

[0021] Fig.10 It is a structural block diagram of a computer terminal according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0023] 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 sequence. 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 an order 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.

[0024] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following explanations:

[0025] Secondary Recolor: It is a color grading technique in video or film post-processing. Different secondary color adjustments are made to different areas in the image, such as adjusting, strengthening, enhancing or weakening certain colors of the background and characters separately to achieve a more refined color grading effect.

[0026] The professional editing software used for coloring images and videos in the related art is complex to operate and has a high threshold for use. For a ten-minute video, it may take several days to complete the arduous coloring task, and the coloring cost is high. Specifically, DaVinci video editing software provides a tool for coloring videos, but it cannot conveniently use deep learning algorithms for automatic coloring. DaVinci video editing software can manually colorize videos globally and regionally. The regional coloring here requires users to manually select areas according to tools such as "circle", "polygon", "curve", and "gradient", which is difficult to implement automatically, reducing the coloring efficiency. At the same time, such a selection area cannot be adaptively adjusted according to the video content object, etc., and the entire video shares the same selection area and coloring parameters by default, which is unreasonable when the video object moves or the lens changes. Although the target tracking function is also provided in the DaVinci video editing software, it is not well integrated with the coloring process, and users cannot finely select the object area and perform targeted coloring.

[0027] In addition, other color grading software such as Photoshop and LightRoom, etc., the core function is to grade the color of a single frame image, and it is impossible to extend the color grading results of a certain frame of the video to other frames in the video footage. Using these color grading software to grade frame by frame will bring a huge workload and it is difficult to ensure the continuity between frames. The current preset color grading filters and end-to-end global color grading AI algorithms cannot meet the professional colorists' needs for fine-grained and customized color grading.

[0028] In summary, the prior art has the technical problems of complex operation process for color adjustment of images and videos, resulting in low color adjustment efficiency and poor color adjustment effect. Currently, no effective solution has been proposed for the above problems.

[0029] Example 1

[0030] According to an embodiment of the present application, a method embodiment of color adjustment processing 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.

[0031] The method embodiment provided in Embodiment 1 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 color adjustment method. Figure 1 As shown, the computer terminal 10 (or mobile device) 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 (Microcontroller Unit, MCU) or a programmable logic device (Field Programmable Gate Array, 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 (I / O interface), a Universal Serial Bus (Universal Serial Bus, USB) port (which may be included as one of the ports of the computer 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 shown.

[0032] 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).

[0033] 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 color adjustment processing method 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, the above-mentioned color adjustment processing method 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.

[0034] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] The display may be, for example, a touch screen liquid crystal display (LCD), which may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0036] It should be noted that, in some optional embodiments, the above Figure 1 The computer device (or mobile device) shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of hardware elements and software elements. It should be noted that Figure 1This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the above-described computer device (or mobile device).

[0037] Under the above operating environment, this application provides Figure 2 The color processing method shown. Figure 2 is a flow chart of a color adjustment method according to Embodiment 1 of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0038] Step S21, determining a color adjustment object, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted;

[0039] Step S22, obtaining a plurality of prediction masks and a plurality of prediction color adjustment parameters corresponding to a plurality of color adjustment areas in the color adjustment object;

[0040] Step S23, using a plurality of predicted color adjustment parameters to perform color adjustment processing on a plurality of predicted masks respectively, to obtain a plurality of initial color adjustment results;

[0041] Step S24, performing fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result.

[0042] The above-mentioned color adjustment objects are images or video materials to be color adjusted. Color adjustment objects can be obtained from online image libraries. Many websites provide free or paid image download services. Users can search and download the required images on these platforms for color adjustment. Color adjustment objects can also be obtained from social media. Many social media platforms have a large number of images and video materials uploaded by users. Users can find the required materials by searching keywords or browsing related topic tags. Color adjustment objects can also be provided by professional photographers or photography agencies. Users can contact these professional photographers or agencies to purchase or obtain their works for color adjustment. Color adjustment objects can also be used for users to use cameras or mobile phones to shoot the images or video materials they need, and then perform color adjustment. It should be noted that no matter which channel the image or video material is obtained from, the user needs to ensure compliance with the use license agreement.

[0043] The above-mentioned multiple color adjustment areas refer to areas in an image or video whose colors and tones can be adjusted separately. These areas can be different parts of the entire image or specific objects or areas in the image. By performing different color adjustment on each area, the color and tones of different parts can be adjusted independently to achieve better visual effects. For example, common color adjustment areas include character skin color, background color, sky color, etc.

[0044] The above-mentioned multiple prediction masks correspond to multiple color adjustment areas, and the multiple prediction color adjustment parameters may include but are not limited to contrast, saturation, hue, color temperature, shadows, midtones, highlights, etc. Among them, contrast refers to the degree of difference between the bright and dark parts of the image. High contrast means that the difference between the bright and dark parts of the image is obvious, while low contrast means that the difference between the bright and dark parts is small; saturation refers to the vividness of the color in the image. High saturation means that the color is bright and full, while low saturation means that the color is relatively gray or light; hue refers to the type or attribute of the color in the image. Hue represents the specific category to which the color belongs, such as red, blue, green, etc.; color temperature refers to the overall hue of the image. Color temperature can make the image appear cool (such as more blue) or warm (such as more yellow); Shadow: refers to the darkest part of the image, which is usually located in the area where light cannot reach and appears darker in tone; Midtone refers to the part of the image between shadow and highlight, which is usually the area with moderate brightness in the image; Highlight refers to the brightest part of the image, which is usually located in the area directly exposed to light and appears brighter in tone.

[0045] When multiple predicted color adjustment parameters are used to perform color adjustment on multiple predicted masks respectively, different initial color adjustment results can be generated according to the predicted color adjustment parameters of different Masks, and then the multiple initial color adjustment results can be linearly superimposed in a parallel manner to achieve fusion processing of the multiple initial color adjustment results, thereby generating the final target color adjustment result.

[0046] Based on the above steps S21 to S24, by determining the image to be colored or the video to be colored, multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas in the image to be colored or the video to be colored are obtained, and then the multiple prediction coloring parameters are used to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results, and finally the multiple initial coloring results are fused to obtain the target coloring result, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the coloring efficiency of images and videos, improving the coloring effect, and reducing the coloring cost, thereby solving the technical problems of low coloring efficiency and poor coloring effect caused by the complicated operation process when coloring images and videos in the related technology.

[0047] Optionally, the color processing method proposed in the embodiment of the present application can be applied, but not limited to, in any application scenario involving color grading of images and videos in the fields of health, science, society, language, and art. The color processing method is not only suitable for beautifying images and videos taken manually by the public, but can also help professional users to enhance the color and adjust the style of media content such as movies and television, thereby improving the efficiency of media production or restoration.

[0048] The color adjustment method in the embodiment of the present application is further introduced below.

[0049] In an optional embodiment, in step S22, obtaining a plurality of prediction masks and a plurality of prediction grading parameters corresponding to a plurality of grading regions in the grading object includes:

[0050] Step S221, using a preset color adjustment area generation method to generate multiple color adjustment areas corresponding to the color adjustment object;

[0051] Step S222 , using a parameter predictor in the target network model to obtain a plurality of prediction masks and a plurality of prediction color adjustment parameters corresponding to a plurality of color adjustment regions.

[0052] The above-mentioned preset color adjustment area generation method includes but is not limited to using saliency detection (SalientDetection), color palette model (Color Palette Model), semantic segmentation (Semantic Segmentation) and other methods to generate multiple color adjustment areas corresponding to the color adjustment object. Specifically, a saliency detection algorithm is used to find salient areas in the image, which usually contain important color adjustment objects; based on the saliency detection results, a color palette model can be used to generate a color palette of the color adjustment object, and the color palette model can extract the main colors according to the distribution and frequency of colors; the image can be segmented into different semantic areas using a semantic segmentation algorithm, and the image can be segmented into areas with semantic information using a pre-trained deep learning model, such as a semantic segmentation network, which can help determine the boundaries of the color adjustment object. According to the results of the saliency detection, the colors of the color palette model and the results of the semantic segmentation, the pixels in the image can be assigned to different color adjustment areas, thereby generating multiple color adjustment areas corresponding to the color adjustment object, and these color adjustment areas can be used for subsequent image editing, image segmentation or other related applications.

[0053] Furthermore, a parameter predictor in the target network model is used to obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas.

[0054] Based on the above optional embodiments, by adopting a preset color adjustment area generation method, multiple color adjustment areas corresponding to the color adjustment object are generated, and then a parameter predictor in the target network model is used to obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas. The color adjustment object can be automatically divided into regions, and the prediction masks and prediction color adjustment parameters of the corresponding color adjustment areas can be quickly obtained, thereby effectively improving the color adjustment efficiency.

[0055] In an optional embodiment, the parameter predictor includes: a feature generation network structure, a mask generation network structure and a parameter generation network structure. In step S222, using the parameter predictor in the target network model to obtain multiple predicted masks and multiple predicted color adjustment parameters corresponding to multiple color adjustment areas includes:

[0056] Step S2221, performing feature coding processing on the color adjustment object based on the feature generation network structure to obtain a feature coding result;

[0057] Step S2222, performing mask generation processing on the feature encoding result based on the mask generation network structure to obtain multiple prediction masks, and performing parameter generation processing on the feature encoding result based on the parameter generation network structure to obtain multiple prediction color adjustment parameters corresponding to the multiple prediction masks.

[0058] Based on the feature generation network structure, the color adjustment object is subjected to feature encoding processing to obtain the feature encoding result, wherein the feature generation network structure may include a backbone network and a bottleneck network. In the target network model, the backbone refers to the main part of the model, which is usually a network composed of convolutional layers and pooling layers, and is responsible for extracting the features of the adjustment object. The backbone network is usually designed based on different tasks and requirements, such as classification, object detection or semantic segmentation.

[0059] The neck network is added after the backbone network, and is usually composed of a series of convolutional layers and pooling layers. The role of the neck network is to further integrate and extract the features extracted by the backbone network to obtain the feature encoding result f(X) to better meet the specific task requirements. The neck network can help reduce the size and number of channels of the feature map, or increase the semantic information of the feature map to improve the performance of the model.

[0060] Figure 3 is a schematic diagram of a color adjustment method according to Example 1 of the present application, such as Figure 3 As shown, based on the mask generation network structure (h map ) performs mask generation processing on the feature encoding result to obtain K prediction masks (Mask), namely M1, M2, ······M K, and generate network structures based on parameters (h arg ) performs parameter generation processing on the feature encoding result to obtain multiple predicted color adjustment parameters corresponding to the K predicted masks, for example, the N predicted color adjustment parameters θ corresponding to M1 1,1 ,·····θ 1,N , M K The corresponding N predicted color parameters θ K,1 ,·····θ K,N Among them, the above parameter generation network structure includes convolution layer (Convolution), group normalization layer (Group norm), linear rectification layer (Relu) and average pooling layer (Avg pool).

[0061] Based on the above optional embodiments, feature encoding processing is performed on the coloring object based on a feature generation network structure to obtain a feature encoding result, and then mask generation processing is performed on the feature encoding result based on a mask generation network structure to obtain multiple prediction masks, and parameter generation processing is performed on the feature encoding result based on a parameter generation network structure to obtain multiple prediction coloring parameters corresponding to the multiple prediction masks. It is possible to quickly and automatically obtain multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas, thereby further improving the coloring efficiency.

[0062] In an optional embodiment, in step S23, multiple prediction color adjustment parameters are used to perform color adjustment processing on multiple prediction masks respectively, and multiple initial color adjustment results are obtained, including:

[0063] Step S231, in a renderer of the target network model, determining a plurality of grading operations to be performed based on a plurality of predicted grading parameters;

[0064] Step S232 , performing color adjustment processing on the multiple prediction masks in parallel according to the color adjustment order relationship of the multiple prediction masks and the multiple color adjustment operations, to obtain multiple initial color adjustment results.

[0065] Specifically, in a parallel node renderer of the target network model, a plurality of grading operations to be performed are determined based on a plurality of predicted grading parameters, and the grading operation may be an increase or decrease in a value of at least one predicted grading parameter.

[0066] The above color adjustment order relationship represents the order relationship during color adjustment. According to the color adjustment order relationship of multiple prediction masks and multiple color adjustment operations, multiple prediction masks are color adjusted in parallel to obtain multiple initial color adjustment results.

[0067] Figure 4 is a schematic diagram of another color adjustment method according to Example 1 of the present application, such as Figure 4As shown, the saturation of the prediction mask M1 is increased by 2.57 to obtain the initial color adjustment result 1, and the midtone of the prediction mask M2 is increased by 0.67 to obtain the initial color adjustment result 2, ..., the prediction mask M k The highlights are lowered by 0.62 to obtain the initial color adjustment result K.

[0068] It should be noted that in the above embodiment, only an example of performing one color adjustment operation is given for each prediction mask. In specific implementation, multiple color adjustment operations can be performed serially on each prediction mask according to the color adjustment requirements, that is, the values ​​of multiple different prediction color adjustment parameters are adjusted. The embodiments of the present application do not constitute a specific limitation.

[0069] Based on the above optional embodiments, in the renderer of the target network model, multiple color adjustment operations to be performed are determined based on multiple predicted color adjustment parameters, and then multiple prediction masks are color adjusted in parallel according to the color adjustment order relationship of the multiple prediction masks and the multiple color adjustment operations, so as to obtain multiple initial color adjustment results at the same time, thereby further improving the color adjustment efficiency.

[0070] In an optional embodiment, the color adjustment method in the embodiment of the present application further includes:

[0071] Step S31, using the initial network model to perform parameter prediction training on the input sample to obtain multiple training masks and multiple training color adjustment parameters, and using the multiple training color adjustment parameters to perform color adjustment processing on the multiple training masks respectively to obtain output samples;

[0072] Step S32, calculating the target loss using the output sample, the real sample, the multiple training masks, and the multiple real masks corresponding to the preset color adjustment area generation method;

[0073] Step S33, updating the model parameters of the initial network model based on the target loss to obtain the target network model.

[0074] Specifically, a group of input samples and real samples are prepared, and the initial network model is used to generate multiple training masks and multiple training color adjustment parameters for each input sample, and multiple training color adjustment parameters are used to adjust the multiple training masks respectively to obtain output samples. The output samples, real samples, multiple training masks, and multiple real masks corresponding to the preset color adjustment area generation method are used to calculate the target loss, and the model parameters of the initial network model are updated based on the target loss. The above process is repeated until the model reaches the preset convergence condition, thereby obtaining the target network model.

[0075] Through the above optional embodiments, a well-trained target network model can be obtained, which can accurately generate multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas after training. At the same time, since the calculation of the target loss takes into account the differences between the output sample and the real sample and between the multiple training masks and the real mask, the trained target network model has good generalization ability and can process a variety of different input samples. Therefore, through the above embodiments, the automatic processing of image color adjustment can be realized, and the accuracy and efficiency of the color adjustment effect can be improved.

[0076] In an optional embodiment, in step S32, the target loss is calculated using the output sample, the real sample, the multiple training masks, and the multiple real masks corresponding to the preset color adjustment area generation method, including:

[0077] Step S321, using the output sample and the real sample to calculate the first loss, and using multiple training masks and multiple real masks to calculate the second loss;

[0078] Step S322, performing weighted sum calculation on the first loss and the second loss to obtain a target loss.

[0079] Specifically, the first loss L is calculated using the output sample and the real sample recon This can be achieved according to the following formula 1:

[0080] L recon =L1(Image generated ,Image groundtruth ) Formula 1

[0081] Among them, Image generated For the output sample, Image groundtruth For real samples.

[0082] The second loss L is calculated using multiple training masks and multiple real masks mask This can be achieved according to the following formula 2:

[0083]

[0084] Among them, Dice_Loss is a loss function commonly used in image segmentation tasks, which measures the similarity between the segmentation results predicted by the model and the actual segmentation map, and CrossEntropy_Loss is the cross entropy loss. Since multiple color-adjusting areas corresponding to the color-adjusting object can be generated by using methods such as Salient Detection, Color Palette Model, and Semantic Segmentation, the second loss L mask It can be determined according to the preset color adjustment area generation method.

[0085] For the first loss L recon With the second loss L mask Perform weighted sum calculation to obtain the target loss L, which can be achieved using the following formula 3:

[0086] L=L recon +λL mask Formula 3

[0087] Among them, λ is used to represent the weight coefficient. Each Mask is a color adjustment area, but this Mask is not a 0-1 binary area, but it can be understood as a matrix with the same size as the image. The value of each position of the matrix is ​​between 0-1. The closer the value is to 1, the greater the weight coefficient for adjusting it. Compared with the 0-1 binary color adjustment area, the advantage of the embodiment of the present application is that the color change at the edge of the object will not be too drastic, that is, smoothing (soft) is performed on the edge of the object.

[0088] Based on the above optional embodiments, the first loss is calculated by using the output sample and the real sample, and the second loss is calculated by using multiple training masks and multiple real masks, and then the first loss and the second loss are weighted summed to obtain the target loss, which is used to train the target network model and improve the performance of the target network model, thereby improving the color adjustment efficiency during color adjustment processing and obtaining a better color adjustment effect.

[0089] In an optional embodiment, the color processing method in the embodiment of the present application also includes: using a preset node graph to describe the color relationship of multiple color adjustment areas, wherein the preset node graph includes: multiple nodes, the multiple nodes are respectively used to record multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas, parent node information and child node information of the multiple nodes, and the sequential relationship between the multiple nodes represents the order of color adjustment.

[0090] The above preset node graph is a color grading node mesh graph, which can be generated based on the input image or video keyframe. Each node (Node) includes the Mask of the corresponding color grading area, the predicted color grading parameters of the area, the parent node information of the node and the next child node information pointed to. The order relationship between nodes represents the order relationship during color grading. If there are multiple nodes of the same level, the initial color grading results of different color grading nodes can be fused to become the final output target color grading result.

[0091] Based on the above optional embodiment, a preset node graph is used to describe the color adjustment relationship of multiple color adjustment areas, which can clarify the color adjustment sequence, achieve orderly color adjustment, and further improve the color adjustment efficiency.

[0092] In an optional embodiment, the color adjustment method in the embodiment of the present application further includes:

[0093] Step 41, in response to the color grading object being a key frame obtained after storyboarding the video, performing region tracking on a plurality of prediction masks and a plurality of prediction color grading parameters corresponding to the key frame, and determining a plurality of prediction masks and a plurality of prediction color grading parameters corresponding to other types of image frames associated with the key frame;

[0094] Step 42 , using a plurality of predicted color adjustment parameters corresponding to the other types of image frames to perform color adjustment processing on a plurality of predicted masks corresponding to the other types of image frames, to obtain color adjustment results corresponding to the other types of image frames.

[0095] In the process of region tracking, an alternative open source tracking model (MiVOS) is adopted in the embodiment of the present application, which can perform region tracking by mask propagation. According to a given object mask, MiVOS tracks the object in the video shot and generates the corresponding object mask in the subsequent process.

[0096] Specifically, according to the multiple prediction masks and multiple prediction color adjustment parameters corresponding to a given key frame, MiVOS tracks the key frame in the video shot, and subsequently generates multiple prediction masks and multiple prediction color adjustment parameters corresponding to the remaining types of image frames associated with the key frame. Among them, the key frame is a frame in the video that has important information, obvious changes, or represents the content of the video. Furthermore, the multiple prediction color adjustment parameters corresponding to the remaining types of image frames are used to perform color adjustment on the multiple prediction masks corresponding to the remaining types of image frames to obtain the color adjustment results corresponding to the remaining types of image frames.

[0097] The embodiment of the present application can provide a color adjustment algorithm software development kit (SDK), which is responsible for receiving the predicted color adjustment parameters passed in by the front end and calling the color adjustment algorithm to process the video frame or image. When processing more than one frame of the image, that is, when performing video color adjustment, all video frames can be passed into the color adjustment algorithm for analysis, the video is split into scenes and key frames are extracted, and the predicted color adjustment parameters and predicted masks of the key frames are obtained. After the analysis of all key frames is completed, the video frame is passed into the color adjustment algorithm again for rendering. When rendering a certain frame, if the current frame is a key frame, the color adjustment algorithm reads the predicted color adjustment parameters corresponding to the key frame and renders it; if the frame is not a key frame, the color adjustment algorithm will perform area tracking according to the key frame corresponding to the frame to determine the color adjustment mask and predicted color adjustment parameters of the frame, and finally render and output to obtain the target color adjustment result.

[0098] When manually adjusting the color of a key frame, the corresponding frame attribute (Frame Attribute) structure file will be modified according to the predicted color adjustment parameters passed in by the front end, and the key frame rendering will be performed again. At the same time, for other frames in the lens corresponding to the key frame, the area tracking and rendering will be re-performed when called. For example, when switching to other frames in the front-end timeline or rendering the entire video, the area tracking and rendering will be re-performed. It should be noted that the embodiment of the present application does not currently support manual color adjustment of frames other than key frames. In addition, the color adjustment algorithm SDK is also compatible with the processing of a single image. When only one frame of the image is processed, the frame is used as the only key frame, and other processing procedures are consistent with the case of more than one frame, so they will not be repeated.

[0099] Based on the above optional embodiments, in response to the color adjustment object being a key frame obtained after storyboarding a video, regional tracking is performed on multiple prediction masks and multiple prediction color adjustment parameters corresponding to the key frame, and multiple prediction masks and multiple prediction color adjustment parameters corresponding to other types of image frames associated with the key frame are determined. Then, the multiple prediction color adjustment parameters corresponding to the other types of image frames are used to perform color adjustment on the multiple prediction masks corresponding to the other types of image frames, and color adjustment results corresponding to the other types of image frames are obtained. This allows adaptive adjustment as the video content changes, avoiding the entire video sharing a selection area or the need to manually adjust the selection area frame by frame, thereby ensuring frame continuity while obtaining good color adjustment results.

[0100] In an optional embodiment, the color processing method in the embodiment of the present application also includes: using a frame attribute structure to store multiple predicted color parameters corresponding to key frames, and using a video attribute structure to store the video's storyboard information, duration information, and key frames corresponding to the storyboard information.

[0101] Specifically, the predicted color adjustment parameters of each key frame are encapsulated into a Frame Attribute structure to be written into a file, and the video's storyboard information, duration information, and key frame information corresponding to each storyboard are saved into a video attribute structure to be written into a file for storage. It should be noted that the Frame Attribute structure and the Video Attribute structure are written into a file instead of being saved in memory in order to avoid memory overflow when the number of frames is large.

[0102] Based on the above optional embodiments, by using a frame attribute structure to store multiple predicted color adjustment parameters corresponding to key frames, and using a video attribute structure to store the storyboard information, duration information, and key frames corresponding to the storyboard information of the video, it is possible to effectively avoid memory overflow and affect the color adjustment processing progress, thereby further improving the color adjustment efficiency.

[0103] In an optional embodiment, a graphical user interface is provided by a terminal device, and the content displayed by the graphical user interface at least partially includes a color adjustment processing scene. The graphical user interface includes: a first touch area, a second touch area, a third touch area, and a fourth touch area. The color adjustment processing method in the embodiment of the present application also includes:

[0104] Step S51, displaying at least one node in a preset node graph in a first touch area;

[0105] Step S52, displaying a prediction mask corresponding to any node in the at least one node in the second touch area;

[0106] Step S53, displaying the predicted color adjustment parameters corresponding to any node in the third touch area;

[0107] Step S54: displaying a preview color adjustment result corresponding to any node in the fourth touch area.

[0108] In the above optional embodiment, the above graphical user interface may be a product interface of a single-frame color grading editor or a product interface of a video color grading editor. The color grading scene displayed in the above graphical user interface may be a color grading scene for beautifying images and videos taken manually by the general public or a color grading scene for professional users to enhance the color and adjust the style of media content such as movies and television.

[0109] The first touch area in the above-mentioned graphical user interface can display at least one node in the color node mesh diagram, the second touch area can display the predicted mask corresponding to any node in the at least one node, the third touch area can display the predicted color parameters corresponding to any node, for example, displaying the color wheel and the predicted color parameter values, and the fourth touch area can display the preview color result corresponding to any node.

[0110] Based on the above optional embodiments, by displaying at least one node in the preset node graph in the first touch area, displaying the predicted mask corresponding to any node in the at least one node in the second touch area, displaying the predicted color adjustment parameters corresponding to any node in the third touch area, and displaying the preview color adjustment result corresponding to any node in the fourth touch area, the color adjustment process can be displayed intuitively, thereby greatly reducing the difficulty of color adjustment for images and videos and further improving the color adjustment efficiency.

[0111] In an optional embodiment, the color adjustment method in the embodiment of the present application further includes:

[0112] Step S55, in response to a first touch operation on a target node in the preset node graph displayed in the first touch area, displaying a prediction mask corresponding to the target node in the second touch area;

[0113] Step S55, in response to the second touch operation on the third touch area, editing the predicted color adjustment parameters corresponding to the target node to obtain edited color adjustment parameters;

[0114] Step S57 , displaying a preview color adjustment result corresponding to the edited color adjustment parameters in the fourth touch area.

[0115] The above-mentioned graphical user interface also includes a first control (or a first touch area). When a first touch operation acting on the first control (or the first touch area) is detected, a prediction mask corresponding to the target node is displayed in a second touch area.

[0116] The above-mentioned graphical user interface also includes a second control (or a third touch area). When a second touch operation acting on the second control (or the third touch area) is detected, the predicted color adjustment parameters corresponding to the target node are edited to obtain the edited color adjustment parameters.

[0117] It should be noted that the first touch operation and the second touch operation can both be operations in which a user touches the display screen of the terminal device with a finger and touches the terminal device. The touch operation can include single-point touch and multi-point touch, wherein the touch operation of each touch point can include click, long press, heavy press, swipe, etc. The first touch operation and the second touch operation can also be touch operations implemented by input devices such as a mouse and a keyboard.

[0118] Based on the above optional embodiments, in response to a first touch operation on a target node in a preset node graph displayed in a first touch area, a prediction mask corresponding to the target node is displayed in a second touch area, and then in response to a second touch operation on a third touch area, the predicted color adjustment parameters corresponding to the target node are edited to obtain the edited color adjustment parameters, and finally, a preview color adjustment result corresponding to the edited color adjustment parameters is displayed in a fourth touch area, thereby achieving flexible adjustment of the preview color adjustment result to better meet the color adjustment needs of users.

[0119] The color processing method proposed in the embodiment of the present application can be applied to a single-frame color editor and a video color editor. The implementation process of color processing using a single-frame color editor and a video color editor is described in detail below.

[0120] The processing steps of the single-frame color editor include: opening the image; automatically analyzing and coloring; manually modifying the color parameters and rendering the results in real time; saving the color parameters and the result image. The single-frame color editor can combine automatic coloring and manual coloring, display key coloring nodes, and facilitate secondary coloring by area.

[0121] After opening a single-frame image, the single-frame color editor first outputs a preview of the color adjustment result obtained by automatic color adjustment based on the color processing method in the embodiment of the present application, and displays the prediction mask and its corresponding prediction color adjustment parameters. Then the single-frame color editor supports manual adjustment of the colors of different areas of the image.

[0122] During the automatic color grading stage, the editor calls the color grading algorithm SDK to analyze the current image, saves the image parameters and predicted color grading parameters, and then calls the color grading algorithm SDK again to read the predicted color grading parameters and render the colored image. The purpose of saving and then reading the parameters here is to be compatible with the multi-frame processing flow to avoid memory overflow.

[0123] In the manual adjustment stage, the smallest adjustment unit is a node, and each color adjustment node corresponds to a different prediction mask and prediction color adjustment parameter. It should be noted that, according to different color adjustment algorithm types, the prediction masks may or may not overlap. Figure 5 is a schematic diagram of a graphical user interface of a color adjustment method according to Example 1 of the present application, such as Figure 5 As shown in the figure, in the product interface of the single-frame color editor, in order to intuitively display the color parameters and segmented areas, four areas can be displayed: color nodes, segmented area masks, color wheels and color parameter values, and result image preview. Among them, the color nodes and segmented area masks intuitively display the logical relationship between different color areas, that is, different areas are adjusted in parallel. Clicking different color nodes in the upper left window will switch the segmented area corresponding to the node in the upper right window, and the color parameters corresponding to the node will be displayed in the lower left window. After selecting the color node corresponding to a segmented area and modifying the color parameters in the control window, the single-frame color editor obtains the front-end color parameters and calls the color algorithm SDK for processing. When editing, the color rendering results can be previewed in real time in the result preview window in the lower right corner. After the color process is completed, the editor can save the result image and the corresponding color parameters as needed, such as keeping structures such as Frame Attribute and Video Attribute.

[0124] The processing steps of the video color editor include: opening the video; automatically analyzing all frames to generate key frame color parameters and rendering the current frame; manually modifying the key frame color parameters and rendering the results in real time; rendering all frames and saving the video. The video color editor can analyze and automatically adjust the video, support manual color adjustment of key frames, and extend the key frame color adjustment parameters to other video frames in the shot segment based on the area tracking algorithm to ensure the color consistency of the video segment.

[0125] The video color editor is integrated into the professional video editor and provides color grading services in the form of Multi-Layer Thin Film Filter (MLT Filter). MLT Filter is an image processing filter, mainly used in film and video production. By adjusting parameters such as color, contrast and saturation, the film has a specific appearance and style, thereby enhancing the artistic sense and viewing quality of the film.

[0126] After opening the video file, you first need to open the MLT Filter, Secondary Recolor, which will call the coloring algorithm SDK to perform a complete analysis and automatic coloring of the video. During this process, the SDK will save the coloring analysis parameters and render the current frame. When switching frames in the timeline, the MLT Filter will obtain the frame number and render the corresponding frame. At the same time, the keyframes generated by the algorithm analysis will be displayed in the keyframe axis. You can switch keyframes and enable the single-frame coloring editor in the coloring editor to color the keyframes. The process here is the same as the single-frame coloring process. After editing, the saved Frame Attribute corresponding to the keyframe will be modified. When calling the frame in the shot, it will be tracked and rendered according to the new keyframe parameters. After editing the keyframes as needed, you can choose to render all frames and save the color-graded video to ensure the color consistency of the video clip.

[0127] Compared with other software in the related art that is difficult to automatically obtain the selected area for automatic secondary color adjustment, the color adjustment processing method in the embodiment of the present application can apply segmentation methods such as saliency detection to the video frame / image to obtain the secondary color adjustment area, and call the color adjustment algorithm to adjust the color of different areas, and support manual adjustment of color adjustment parameters. At the same time, the area tracking algorithm of the color adjustment processing method in the embodiment of the present application ensures that the color adjustment area can be adaptively adjusted as the video content changes, avoiding the entire video sharing a selection area or the need to manually adjust the selection area frame by frame.

[0128] Compared with other software in the related art, which has difficulty in handling the color grading requirements of long videos and cannot ensure inter-frame continuity while obtaining good color grading results, the self-developed editor in the embodiment of this application can call the color grading SDK to analyze the video or image, automatically segment the video footage, obtain key frames, and apply segmentation methods such as saliency detection to the key frames to obtain the color grading area and perform color grading; for non-key frames, the embodiment of this application can apply the area tracking algorithm to expand the key frame selection area and color grading parameters to ensure inter-frame continuity after color grading. Through the above method, the embodiment of this application supports manual editing and greatly reduces the difficulty of video color grading on the basis of obtaining good automatic color grading results.

[0129] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0130] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0131] Through the description of the above implementation methods, those skilled in the art can clearly understand that the 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. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), including several instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0132] Example 2

[0133] According to an embodiment of the present application, a color adjustment method is also provided. Figure 6 is a flow chart of a color adjustment method according to Embodiment 2 of the present application. Figure 6 As shown, the method comprises the following steps:

[0134] Step S61, receiving a color adjustment object from a client, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted;

[0135] Step S62, obtaining a plurality of prediction masks and a plurality of prediction grading parameters corresponding to a plurality of grading regions in the grading object, performing grading processing on the plurality of prediction masks using the plurality of prediction grading parameters to obtain a plurality of initial grading results, and performing fusion processing on the plurality of initial grading results to obtain a target grading result;

[0136] Step S63: Feedback the target color adjustment result to the client.

[0137] Based on the above steps S61 to S63, by receiving a color adjustment object from the client, wherein the color adjustment object is an image to be colored or a video to be colored, multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object are obtained, and multiple prediction color adjustment parameters are used to perform color adjustment on the multiple prediction masks respectively to obtain multiple initial color adjustment results, and the multiple initial color adjustment results are fused to obtain a target color adjustment result, and finally the target color adjustment result is fed back to the client, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the color adjustment efficiency of images and videos, improving the color adjustment effect, and reducing the color adjustment cost, thereby solving the technical problems of low color adjustment efficiency and poor color adjustment effect caused by the complicated operation process when coloring images and videos in the related technology.

[0138] Optionally, the color processing method proposed in the embodiment of the present application can be applied, but not limited to, in any application scenario involving color grading of images and videos in the fields of health, science, society, language, and art. The color processing method is not only suitable for beautifying images and videos taken manually by the public, but can also help professional users to enhance the color and adjust the style of media content such as movies and television, thereby improving the efficiency of media production or restoration.

[0139] Figure 7 is a schematic diagram of a color adjustment method performed on a cloud server according to Embodiment 2 of the present application, such as Figure 7 As shown, the client uploads the image to be colored or the video to be colored to the cloud server, and the cloud server obtains multiple prediction masks and multiple prediction coloring parameters corresponding to the multiple coloring areas in the coloring object, uses the multiple prediction coloring parameters to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results, and fuses the multiple initial coloring results to obtain the target coloring result. Then, the cloud server will feedback the target coloring result to the above-mentioned client, and the final target coloring result will be displayed to the user through the client's graphical user interface.

[0140] It should be noted that the above-mentioned color processing method provided in the embodiment of the present invention can be but is not limited to application scenarios for color grading of images and videos. Through the interaction between the SaaS server and the client, multiple prediction masks and multiple prediction color parameters corresponding to multiple color grading areas in the color grading object are obtained, and multiple prediction color parameters are used to perform color grading on multiple prediction masks respectively to obtain multiple initial color grading results, and multiple initial color grading results are fused to obtain a target color grading result, and the returned target color grading result is provided to the customer through the client.

[0141] For the parts not described in detail in the above embodiments of the present application, please refer to the relevant description of Embodiment 1 and will not be described in detail here.

[0142] Example 3

[0143] According to an embodiment of the present application, a color adjustment processing device for implementing the above color adjustment processing method is also provided. Figure 8 is a structural block diagram of a color adjustment processing device according to Embodiment 1 of the present application, such as Figure 8 As shown, the device comprises:

[0144] A determination module 801 is used to determine a color adjustment object, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted;

[0145] An acquisition module 802 is used to acquire a plurality of prediction masks and a plurality of prediction color adjustment parameters corresponding to a plurality of color adjustment areas in a color adjustment object;

[0146] A color adjustment module 803 is used to perform color adjustment processing on multiple prediction masks respectively using multiple prediction color adjustment parameters to obtain multiple initial color adjustment results;

[0147] The fusion module 804 is used to fuse the multiple initial color adjustment results to obtain a target color adjustment result.

[0148] Optionally, the acquisition module 802 is further used to: generate multiple color adjustment areas corresponding to the color adjustment object using a preset color adjustment area generation method; and obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas using a parameter predictor in the target network model.

[0149] Optionally, the acquisition module 802 is also used to: perform feature encoding processing on the color adjustment object based on the feature generation network structure to obtain a feature encoding result; perform mask generation processing on the feature encoding result based on the mask generation network structure to obtain multiple prediction masks; and perform parameter generation processing on the feature encoding result based on the parameter generation network structure to obtain multiple prediction color adjustment parameters corresponding to the multiple prediction masks.

[0150] Optionally, the color adjustment module 803 is also used to: in the renderer of the target network model, determine multiple color adjustment operations to be performed based on multiple predicted color adjustment parameters; perform parallel color adjustment processing on multiple predicted masks according to the color adjustment order relationship of the multiple predicted masks and the multiple color adjustment operations to obtain multiple initial color adjustment results.

[0151] Optionally, the color processing device also includes: a training module 805, which is used to use the initial network model to perform parameter prediction training on the input sample to obtain multiple training masks and multiple training color adjustment parameters, and use multiple training color adjustment parameters to perform color adjustment processing on multiple training masks respectively to obtain output samples; a calculation module 806, which is used to calculate the target loss using the output sample, the real sample, the multiple training masks and the multiple real masks corresponding to the preset color adjustment area generation method; an update module 807, which is used to update the model parameters of the initial network model based on the target loss to obtain the target network model.

[0152] Optionally, the calculation module 806 is further used to: calculate a first loss using the output sample and the real sample, and calculate a second loss using multiple training masks and multiple real masks; and perform weighted sum calculation on the first loss and the second loss to obtain a target loss.

[0153] Optionally, the color processing device also includes: a processing module 808, which is used to use a preset node graph to describe the color relationship of multiple color adjustment areas, wherein the preset node graph includes: multiple nodes, the multiple nodes are used to record multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas, parent node information and child node information of the multiple nodes, and the sequential relationship between the multiple nodes represents the order of color adjustment.

[0154] Optionally, the processing module 808 is also used for: in response to the color adjustment object being a key frame obtained after storyboarding the video, performing area tracking on multiple prediction masks and multiple prediction color adjustment parameters corresponding to the key frame, and determining multiple prediction masks and multiple prediction color adjustment parameters corresponding to other types of image frames associated with the key frame; and performing color adjustment processing on multiple prediction masks corresponding to other types of image frames using multiple prediction color adjustment parameters corresponding to other types of image frames, to obtain color adjustment results corresponding to other types of image frames.

[0155] Optionally, the color processing device also includes: a storage module 809, which is used to store multiple predicted color parameters corresponding to key frames using a frame attribute structure, and to store the video's storyboard information, duration information, and key frames corresponding to the storyboard information using a video attribute structure.

[0156] Optionally, the color processing device also includes: a display module 810, which is used to display at least one node in a preset node graph in a first touch area; display a predicted mask corresponding to any node in at least one node in a second touch area; display a predicted color adjustment parameter corresponding to any node in a third touch area; and display a preview color adjustment result corresponding to any node in a fourth touch area.

[0157] Optionally, the display module 810 is also used to respond to a first touch operation on a target node in a preset node graph displayed in the first touch area, and display a predicted mask corresponding to the target node in the second touch area; the processing module 808 is also used to respond to a second touch operation on the third touch area, edit the predicted color adjustment parameters corresponding to the target node, and obtain the edited color adjustment parameters; the display module 810 is also used to display a preview color adjustment result corresponding to the edited color adjustment parameters in the fourth touch area.

[0158] By adopting the embodiment of the present application, an image to be colored or a video to be colored is determined, and then multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas in the image to be colored or the video to be colored are obtained. Then, multiple prediction coloring parameters are used to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results. Finally, the multiple initial coloring results are fused to obtain a target coloring result, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the coloring efficiency of images and videos, improving the coloring effect, and reducing the coloring cost, thereby solving the technical problems of low coloring efficiency and poor coloring effect caused by the complicated operation process when coloring images and videos in the related art.

[0159] It should be noted that the above-mentioned determination module 801, acquisition module 802, color adjustment module 803 and fusion module 804 correspond to steps S21 to S24 in Example 1, and the four 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 or units may 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), and the above-mentioned modules may also be part of the device and may be run in the computer terminal 10 provided in Example 1.

[0160] Fig. 9 is a structural block diagram of a color processing device according to Embodiment 2 of the present application, such as Fig. 9 As shown, the device comprises:

[0161] The receiving module 901 is used to receive a color adjustment object from a client, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted;

[0162] The processing module 902 is used to obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object, use the multiple prediction color adjustment parameters to perform color adjustment processing on the multiple prediction masks respectively to obtain multiple initial color adjustment results, and perform fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result;

[0163] The feedback module 903 is used to feed back the target color adjustment result to the client.

[0164] By adopting the embodiment of the present application, a color adjustment object is received from a client, wherein the color adjustment object is an image to be colored or a video to be colored, and then multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object are obtained, and multiple prediction color adjustment parameters are used to perform color adjustment on the multiple prediction masks respectively to obtain multiple initial color adjustment results, and the multiple initial color adjustment results are fused to obtain a target color adjustment result, and finally the target color adjustment result is fed back to the client, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the color adjustment efficiency of images and videos, improving the color adjustment effect, and reducing the color adjustment cost, thereby solving the technical problems of low color adjustment efficiency and poor color adjustment effect caused by the complicated operation process when coloring images and videos in the related technology.

[0165] It should be noted that the receiving module 901, the processing module 902 and the feedback module 903 correspond to steps S61 to S63 in Example 2, 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 2. It should be noted that the above-mentioned modules or units may 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), and the above-mentioned modules may also be part of the device and may be run in the computer terminal 10 provided in Example 1.

[0166] Example 4

[0167] The embodiment of the present application may provide a computer terminal, which may be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced by a terminal device such as a mobile terminal.

[0168] 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.

[0169] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the color processing method: determine the color object, wherein the color object is an image to be colored or a video to be colored; obtain multiple prediction masks and multiple prediction color parameters corresponding to multiple color areas in the color object; use multiple prediction color parameters to perform color processing on multiple prediction masks respectively to obtain multiple initial color results; and perform fusion processing on the multiple initial color results to obtain a target color result.

[0170] Optionally, Fig.10 is a structural block diagram of a computer terminal according to an embodiment of the present application. Fig.10 As shown, the computer terminal may include: one or more (only one is shown in the figure) processors 102, a memory 104, a storage controller, and a peripheral interface, wherein the peripheral interface is connected to a radio frequency module, an audio module, and a display.

[0171] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the color adjustment processing method and device in the embodiment 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, realizing the above-mentioned color adjustment 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 can be connected to the computer 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.

[0172] The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: determine a color adjustment object, wherein the color adjustment object is an image to be colored or a video to be colored; obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object; use multiple prediction color adjustment parameters to perform color adjustment on multiple prediction masks respectively to obtain multiple initial color adjustment results; and perform fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result.

[0173] Optionally, the processor may also execute the program code of the following steps: using a preset color adjustment area generation method to generate multiple color adjustment areas corresponding to the color adjustment object; using a parameter predictor in the target network model to obtain multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas.

[0174] Optionally, the processor may also execute the program code of the following steps: performing feature encoding processing on the color adjustment object based on the feature generation network structure to obtain a feature encoding result; performing mask generation processing on the feature encoding result based on the mask generation network structure to obtain multiple prediction masks; and performing parameter generation processing on the feature encoding result based on the parameter generation network structure to obtain multiple prediction color adjustment parameters corresponding to the multiple prediction masks.

[0175] Optionally, the processor may also execute the program code of the following steps: in the renderer of the target network model, determining multiple color adjustment operations to be performed based on multiple predicted color adjustment parameters; performing parallel color adjustment processing on multiple predicted masks according to the color adjustment order relationship of multiple predicted masks and multiple color adjustment operations to obtain multiple initial color adjustment results.

[0176] Optionally, the processor may also execute the program code of the following steps: using the initial network model to perform parameter prediction training on the input sample to obtain multiple training masks and multiple training color adjustment parameters, and using multiple training color adjustment parameters to perform color adjustment on multiple training masks respectively to obtain output samples; using the output sample, the real sample, multiple training masks and multiple real masks corresponding to the preset color adjustment area generation method to calculate the target loss; based on the target loss, the model parameters of the initial network model are updated to obtain the target network model.

[0177] Optionally, the processor may also execute the following program code: using the output sample and the real sample to calculate the first loss, and using multiple training masks and multiple real masks to calculate the second loss; performing weighted sum calculation on the first loss and the second loss to obtain the target loss.

[0178] Optionally, the processor may also execute the program code of the following steps: using a preset node graph to describe the color adjustment relationship of multiple color adjustment areas, wherein the preset node graph includes: multiple nodes, the multiple nodes are respectively used to record multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas, parent node information and child node information of the multiple nodes, and the sequential relationship between the multiple nodes represents the order of color adjustment.

[0179] Optionally, the processor may also execute the program code of the following steps: in response to the color adjustment object being a key frame obtained after storyboarding the video, performing area tracking on multiple prediction masks and multiple prediction color adjustment parameters corresponding to the key frame, and determining multiple prediction masks and multiple prediction color adjustment parameters corresponding to other types of image frames associated with the key frame; performing color adjustment processing on multiple prediction masks corresponding to other types of image frames using multiple prediction color adjustment parameters corresponding to other types of image frames, and obtaining color adjustment results corresponding to other types of image frames.

[0180] Optionally, the processor may also execute the following program code steps: using a frame attribute structure to store multiple predicted color adjustment parameters corresponding to key frames, and using a video attribute structure to store the video's storyboard information, duration information, and key frames corresponding to the storyboard information.

[0181] Optionally, the processor may also execute program codes of the following steps: displaying at least one node in a preset node graph in a first touch area; displaying a predicted mask corresponding to any node in at least one node in a second touch area; displaying a predicted color adjustment parameter corresponding to any node in a third touch area; and displaying a preview color adjustment result corresponding to any node in a fourth touch area.

[0182] Optionally, the processor may also execute program codes of the following steps: in response to a first touch operation on a target node in a preset node graph displayed in a first touch area, display a prediction mask corresponding to the target node in a second touch area; in response to a second touch operation on a third touch area, edit the prediction color parameters corresponding to the target node to obtain edited color parameters; and display a preview color result corresponding to the edited color parameters in a fourth touch area.

[0183] Optionally, the processor may also execute program code of the following steps: receiving a color adjustment object from a client, wherein the color adjustment object is an image to be colored or a video to be colored; obtaining multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object, using multiple prediction color adjustment parameters to perform color adjustment on multiple prediction masks respectively to obtain multiple initial color adjustment results, and fusing the multiple initial color adjustment results to obtain a target color adjustment result; and feeding back the target color adjustment result to the client.

[0184] By adopting the embodiment of the present application, an image to be colored or a video to be colored is determined, and then multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas in the image to be colored or the video to be colored are obtained. Then, multiple prediction coloring parameters are used to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results. Finally, the multiple initial coloring results are fused to obtain a target coloring result, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the coloring efficiency of images and videos, improving the coloring effect, and reducing the coloring cost, thereby solving the technical problems of low coloring efficiency and poor coloring effect caused by the complicated operation process when coloring images and videos in the related art.

[0185] It can be understood by those skilled in the art that Fig.10 The structure shown is for illustration only, and the computer terminal 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, and a mobile Internet device (Mobile Internet Devices, MID). Fig.10 It does not limit the structure of the above electronic device. For example, the computer terminal may also include Fig.10More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Fig.10 Different configurations shown.

[0186] 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.

[0187] Example 5

[0188] The embodiment of the present application further provides a computer-readable storage medium. Optionally, in this embodiment, the computer-readable storage medium can be used to store the program code executed by the color adjustment method provided in the first embodiment.

[0189] 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.

[0190] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: determining a color adjustment object, wherein the color adjustment object is an image to be colored or a video to be colored; obtaining multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object; using multiple prediction color adjustment parameters to perform color adjustment processing on multiple prediction masks respectively to obtain multiple initial color adjustment results; and fusing the multiple initial color adjustment results to obtain a target color adjustment result.

[0191] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: using a preset color area generation method to generate multiple color areas corresponding to the color object; using a parameter predictor in the target network model to obtain multiple prediction masks and multiple prediction color parameters corresponding to the multiple color areas.

[0192] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: performing feature encoding processing on the color adjustment object based on the feature generation network structure to obtain a feature encoding result; performing mask generation processing on the feature encoding result based on the mask generation network structure to obtain multiple prediction masks, and performing parameter generation processing on the feature encoding result based on the parameter generation network structure to obtain multiple prediction color adjustment parameters corresponding to the multiple prediction masks.

[0193] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: in a renderer of a target network model, determining multiple color adjustment operations to be performed based on multiple predicted color adjustment parameters; performing parallel color adjustment processing on multiple prediction masks according to the color adjustment order relationship of multiple prediction masks and multiple color adjustment operations to obtain multiple initial color adjustment results.

[0194] Optionally, in this embodiment, the storage medium is configured to store a program code for executing the following steps: using the initial network model to perform parameter prediction training on the input sample to obtain multiple training masks and multiple training color adjustment parameters, and using the multiple training color adjustment parameters to perform color adjustment processing on the multiple training masks respectively to obtain output samples; using the output sample, the real sample, the multiple training masks and the multiple real masks corresponding to the preset color adjustment area generation method to calculate the target loss; based on the target loss, the model parameters of the initial network model are updated to obtain the target network model.

[0195] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps: using the output sample and the real sample to calculate the first loss, and using multiple training masks and multiple real masks to calculate the second loss; performing weighted sum calculation on the first loss and the second loss to obtain the target loss.

[0196] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: using a preset node graph to describe the color adjustment relationship of multiple color adjustment areas, wherein the preset node graph includes: multiple nodes, the multiple nodes are respectively used to record multiple prediction masks and multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas, parent node information and child node information of the multiple nodes, and the sequential relationship between the multiple nodes represents the order of color adjustment.

[0197] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: in response to the color grading object being a key frame obtained after storyboarding a video, performing area tracking on multiple prediction masks and multiple prediction color grading parameters corresponding to the key frame, and determining multiple prediction masks and multiple prediction color grading parameters corresponding to other types of image frames associated with the key frame; and performing color grading processing on multiple prediction masks corresponding to other types of image frames using multiple prediction color grading parameters corresponding to other types of image frames to obtain color grading results corresponding to other types of image frames.

[0198] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: storing multiple predicted color adjustment parameters corresponding to key frames using a frame attribute structure, and storing the storyboard information, duration information, and key frames corresponding to the storyboard information of the video using a video attribute structure.

[0199] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: displaying at least one node in a preset node graph in a first touch area; displaying a prediction mask corresponding to any node in at least one node in a second touch area; displaying a prediction color adjustment parameter corresponding to any node in a third touch area; and displaying a preview color adjustment result corresponding to any node in a fourth touch area.

[0200] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: in response to a first touch operation on a target node in a preset node graph displayed in a first touch area, displaying a prediction mask corresponding to the target node in a second touch area; in response to a second touch operation on a third touch area, editing the predicted color adjustment parameters corresponding to the target node to obtain edited color adjustment parameters; and displaying a preview color adjustment result corresponding to the edited color adjustment parameters in a fourth touch area.

[0201] Optionally, in this embodiment, the storage medium is configured to store program code for executing the following steps: receiving a color adjustment object from a client, wherein the color adjustment object is an image to be colored or a video to be colored; obtaining multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object, using multiple prediction color adjustment parameters to perform color adjustment on the multiple prediction masks respectively to obtain multiple initial color adjustment results, and fusing the multiple initial color adjustment results to obtain a target color adjustment result; and feeding back the target color adjustment result to the client.

[0202] By adopting the embodiment of the present application, an image to be colored or a video to be colored is determined, and then multiple prediction masks and multiple prediction coloring parameters corresponding to multiple coloring areas in the image to be colored or the video to be colored are obtained. Then, multiple prediction coloring parameters are used to perform coloring processing on the multiple prediction masks respectively to obtain multiple initial coloring results. Finally, the multiple initial coloring results are fused to obtain a target coloring result, thereby achieving the purpose of quickly coloring the image or video to be colored, thereby achieving the technical effect of improving the coloring efficiency of images and videos, improving the coloring effect, and reducing the coloring cost, thereby solving the technical problems of low coloring efficiency and poor coloring effect caused by the complicated operation process when coloring images and videos in the related art.

[0203] 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.

[0204] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0205] 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.

[0206] 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.

[0207] In addition, each functional unit in each embodiment of the present application 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.

[0208] 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, 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 to enable 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 application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0209] The above is only a preferred implementation 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 color adjustment method, characterized in that: include: Determining a color adjustment object, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted; Acquire a plurality of prediction masks and a plurality of prediction grading parameters corresponding to a plurality of grading regions in the grading object; Using the multiple predicted color adjustment parameters to perform color adjustment processing on the multiple predicted masks respectively to obtain multiple initial color adjustment results; The multiple initial color adjustment results are fused to obtain a target color adjustment result.

2. The color adjustment method according to claim 1, characterized in that: Acquiring the plurality of prediction masks and the plurality of prediction color adjustment parameters corresponding to the plurality of color adjustment areas in the color adjustment object includes: Using a preset color adjustment area generation method to generate the multiple color adjustment areas corresponding to the color adjustment object; The parameter predictor in the target network model is used to obtain the multiple prediction masks and the multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas.

3. The color adjustment method according to claim 2, characterized in that: The parameter predictor includes: a feature generation network structure, a mask generation network structure and a parameter generation network structure, and using the parameter predictor in the target network model to obtain the multiple predicted masks and the multiple predicted color adjustment parameters corresponding to the multiple color adjustment areas includes: Performing feature encoding processing on the color adjustment object based on the feature generation network structure to obtain a feature encoding result; The feature encoding result is subjected to mask generation processing based on the mask generation network structure to obtain the multiple prediction masks, and the feature encoding result is subjected to parameter generation processing based on the parameter generation network structure to obtain the multiple prediction color adjustment parameters corresponding to the multiple prediction masks.

4. The color matching method according to claim 2, characterized in that: Using the multiple predicted color adjustment parameters to perform color adjustment processing on the multiple predicted masks respectively to obtain the multiple initial color adjustment results includes: In a renderer of the target network model, determining a plurality of grading operations to be performed based on the plurality of predicted grading parameters; According to the color adjustment order relationship of the multiple prediction masks and the multiple color adjustment operations, the multiple prediction masks are color adjusted in parallel to obtain the multiple initial color adjustment results.

5. The color adjustment method according to claim 2, characterized in that: The color adjustment method also includes: Using the initial network model to perform parameter prediction training on the input sample to obtain a plurality of training masks and the plurality of training color adjustment parameters, and using the plurality of training color adjustment parameters to perform color adjustment processing on the plurality of training masks respectively to obtain output samples; Calculating a target loss using the output sample, the real sample, the multiple training masks, and the multiple real masks corresponding to the preset color adjustment area generation method; The model parameters of the initial network model are updated based on the target loss to obtain the target network model.

6. The color adjustment method according to claim 5, characterized in that: Calculating the target loss using the output sample, the real sample, the multiple training masks, and the multiple real masks corresponding to the preset color adjustment area generation method includes: The output sample and the real sample are used to calculate a first loss, and the multiple training masks and the multiple real masks are used to calculate a second loss; A weighted sum is calculated for the first loss and the second loss to obtain the target loss.

7. The color adjustment method according to claim 1, characterized in that: The color adjustment method also includes: A preset node graph is used to describe the color adjustment relationship of the multiple color adjustment areas, wherein the preset node graph includes: multiple nodes, the multiple nodes are respectively used to record the multiple prediction masks and the multiple prediction color adjustment parameters corresponding to the multiple color adjustment areas, the parent node information and the child node information of the multiple nodes, and the sequential relationship between the multiple nodes represents the order of color adjustment.

8. The color adjustment method according to claim 2, characterized in that: The color adjustment method also includes: In response to the color grading object being a key frame obtained after storyboarding the video, performing region tracking on a plurality of prediction masks and a plurality of prediction color grading parameters corresponding to the key frame, and determining a plurality of prediction masks and a plurality of prediction color grading parameters corresponding to other types of image frames associated with the key frame; The multiple prediction color adjustment parameters corresponding to the other types of image frames are used to perform color adjustment processing on the multiple prediction masks corresponding to the other types of image frames to obtain color adjustment results corresponding to the other types of image frames.

9. The color adjustment method according to claim 8, characterized in that: The color adjustment method also includes: A plurality of predicted color adjustment parameters corresponding to the key frames are stored in a frame attribute structure, and the storyboard information, duration information, and key frames corresponding to the storyboard information of the video are stored in a video attribute structure.

10. The color adjustment method according to claim 7, characterized in that: A graphical user interface is provided through a terminal device, and the content displayed by the graphical user interface at least partially includes a color adjustment processing scene, the graphical user interface includes: a first touch area, a second touch area, a third touch area and a fourth touch area, and the color adjustment processing method further includes: Displaying at least one node in the preset node graph in the first touch area; Displaying a prediction mask corresponding to any node of the at least one node in the second touch area; Displaying the predicted color adjustment parameter corresponding to any node in the third touch area; A preview color adjustment result corresponding to any node is displayed in the fourth touch area.

11. The color matching method according to claim 10, characterized in that: The color adjustment method also includes: In response to a first touch operation on a target node in the preset node graph displayed in the first touch area, displaying a prediction mask corresponding to the target node in the second touch area; In response to a second touch operation on the third touch area, the predicted color adjustment parameter corresponding to the target node is edited to obtain an edited color adjustment parameter; A preview color adjustment result corresponding to the edited color adjustment parameter is displayed in the fourth touch area.

12. A color adjustment method, characterized in that: include: Receiving a color adjustment object from a client, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted; Acquire multiple prediction masks and multiple prediction color adjustment parameters corresponding to multiple color adjustment areas in the color adjustment object, perform color adjustment processing on the multiple prediction masks respectively using the multiple prediction color adjustment parameters to obtain multiple initial color adjustment results, and perform fusion processing on the multiple initial color adjustment results to obtain a target color adjustment result; The target color adjustment result is fed back to the client.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the color adjustment processing method according to any one of claims 1 to 12.

14. An electronic device, characterized in that: include: processor; as well as A memory, connected to the processor, configured to provide the processor with instructions for processing the following processing steps: Determining a color adjustment object, wherein the color adjustment object is an image to be color adjusted or a video to be color adjusted; Acquire a plurality of prediction masks and a plurality of prediction grading parameters corresponding to a plurality of grading regions in the grading object; Using the multiple predicted color adjustment parameters to perform color adjustment processing on the multiple predicted masks respectively to obtain multiple initial color adjustment results; The multiple initial color adjustment results are fused to obtain a target color adjustment result.

Images