A method and system for realizing intelligent rendering of animation pictures

Abstract

The present invention relates to the field of animation rendering, and proposes an intelligent rendering method for realizing animation pictures, including: firstly collecting initial animation pictures, identifying their scenes, analyzing the animation effect requirements and combining the style characteristics to analyze the pictures to obtain scene picture elements, detecting the element type to determine the degree of special effect requirements, thereby determining the special effect processing method and calculating the index, generating a special effect rendering scheme, querying the adaptive color mode, evaluating the visual atmosphere parameters to calculate the color adjustment amount, constructing a color optimization strategy, analyzing the picture layout to determine the perspective type and parameters, formulating a perspective adjustment plan, determining the rendering target based on the special effect rendering scheme, the color optimization strategy and the perspective adjustment plan, optimizing the picture records accordingly, monitoring the rendering quality, and finally generating an intelligent rendering scheme. The present invention can improve the efficiency and quality of animation picture rendering.

Description

Technical Field

[0001] The present invention relates to the field of animation rendering, and in particular to an intelligent rendering method and system for realizing animation pictures. Background Art

[0002] In today's animation rendering field, as audiences' requirements for the quality of animation works continue to rise and animation content creation becomes increasingly prosperous, the rendering effect of animation images has become one of the key factors that determine the success or failure of a work. It occupies a core position in the production process of various animation works (including TV animation, online animation, animated films, etc.).

[0003] At present, the traditional animation rendering method mainly relies on manual operation and conventional software tools. In this process, animators need to manually set a large number of parameters, such as lighting effects, color modes, material textures, etc., to achieve the ideal picture effect. At the same time, the rendering process usually takes a long time, especially for high-quality, high-resolution animation pictures, which consumes a lot of computing resources and time costs. This method has obvious limitations. On the one hand, the manual setting of parameters is inefficient. It is difficult to complete the rendering of complex scenes and a large number of pictures in a short time, which prolongs the animation production cycle. On the other hand, manual operation depends on the experience and skill level of the animator. Different animators may produce different rendering effects. There is a lack of unified high-quality standards, and human errors are prone to occur, which affects the picture quality. Therefore, a method for rendering animation pictures based on intelligent algorithms is needed to improve the efficiency and quality of animation picture rendering. Summary of the invention

[0004] The present invention provides a method and system for realizing intelligent rendering of animation pictures, the main purpose of which is to improve the efficiency and quality of animation picture rendering.

[0005] To achieve the above object, the present invention provides a method for realizing intelligent rendering of animation pictures, comprising:

[0006] Collecting an initial animation picture, identifying an animation scene corresponding to the initial animation picture, analyzing the animation effect requirements required for the animation scene, and performing picture analysis on the initial animation picture based on the animation effect requirements and the style characteristics corresponding to the animation scene to obtain scene picture elements;

[0007] Querying the element type of the scene screen element, identifying the degree of special effect requirement in the element type, determining the special effect processing method corresponding to the initial animation screen based on the special effect requirement, calculating the special effect processing index corresponding to the special effect processing method, and generating the special effect rendering scheme corresponding to the animation scene based on the special effect processing index;

[0008] Querying the color mode that the animation scene can adapt to, evaluating the visual atmosphere parameters of the animation scene, calculating the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and constructing a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount;

[0009] Performing composition analysis on the screen layout of the animation scene to obtain composition analysis data, determining the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, querying the perspective parameters corresponding to the perspective type, and formulating the perspective adjustment plan of the animation scene based on the perspective parameters;

[0010] Based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, determine the rendering target corresponding to the initial animation picture, based on the rendering target, optimize the picture of the animation scene to obtain a picture optimization record, based on the picture optimization record, monitor the rendering quality corresponding to the animation scene, and based on the rendering quality, generate an intelligent rendering scheme corresponding to the initial animation picture.

[0011] Optionally, based on the animation effect requirement and in combination with the style characteristics corresponding to the animation scene, the initial animation picture is analyzed to obtain scene picture elements, including:

[0012] Deconstructing the animation effect requirement to obtain a deconstructed feature set;

[0013] Identifying visual tags in the style characteristics corresponding to the animation scene;

[0014] Based on the deconstructed feature set and the visual label, construct a mapping rule corresponding to the initial animation picture;

[0015] Based on the mapping rule, the initial animation screen is divided into regions to obtain screen division regions;

[0016] Extracting regional visual points in the divided areas of the picture;

[0017] Based on the regional visual points, the initial animation picture is analyzed to obtain scene picture elements.

[0018] Optionally, determining a special effects processing method corresponding to the initial animation picture based on the special effects requirement degree includes:

[0019] Quantitatively grading the degree of special effect demand to obtain a special effect demand level;

[0020] Query the special effect type template corresponding to the special effect requirement level;

[0021] Analyzing the style attributes of the elements in the initial animation picture to obtain style feature information;

[0022] Fusing the special effect type template with the style feature information to obtain a special effect fusion template;

[0023] Extracting key special effect parameters in the special effect fusion template;

[0024] Based on the key special effect parameters, a special effect processing method corresponding to the initial animation picture is determined.

[0025] Optionally, constructing a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount includes:

[0026] Extracting features of the color pattern to obtain pattern feature information;

[0027] Querying a color rule set that matches the pattern feature information;

[0028] Integrating the color adjustment amount with the color rule set to obtain integrated color data;

[0029] Analyzing a color optimization direction corresponding to the integrated color data;

[0030] Based on the color optimization direction, a color optimization strategy corresponding to the initial animation picture is constructed.

[0031] Optionally, performing composition analysis on the picture layout of the animation scene to obtain composition analysis data includes:

[0032] Performing image segmentation processing on the animation scene to obtain segmented image areas;

[0033] identifying key compositional elements in the segmented image region;

[0034] Classify and sort the key composition elements to obtain classified composition elements;

[0035] Analyzing layout features corresponding to the classified composition elements;

[0036] Based on the layout features, a composition analysis is performed on the picture layout of the animation scene to obtain composition analysis data.

[0037] Optionally, formulating a perspective adjustment plan for the animation scene based on the perspective parameter includes:

[0038] Analyzing picture features corresponding to the viewing angle parameters;

[0039] Determining adjustment demand points corresponding to the picture features;

[0040] Based on the adjustment demand point, planning a preliminary adjustment direction corresponding to the animation scene;

[0041] Performing a simulated viewing angle adjustment on the preliminary adjustment direction to obtain a simulated adjustment parameter;

[0042] Based on the simulation adjustment parameters, a viewing angle adjustment plan for the animation scene is formulated.

[0043] Optionally, the performing picture optimization on the animation scene based on the rendering target to obtain a picture optimization record includes:

[0044] Analyze the screen element requirements corresponding to the rendering target;

[0045] Based on the picture element requirements, extracting scene elements in the animation scene;

[0046] Querying the element status corresponding to the scene element;

[0047] Analyze the state optimization point corresponding to the element state;

[0048] Based on the state optimization point, the animation scene is optimized to obtain a picture optimization record.

[0049] Optionally, in order to solve the above problem, the present invention provides an intelligent rendering system for realizing animation pictures, the system comprising:

[0050] A picture analysis module is used to collect an initial animation picture, identify an animation scene corresponding to the initial animation picture, analyze the animation effect requirements required by the animation scene, and perform picture analysis on the initial animation picture based on the animation effect requirements and the style characteristics corresponding to the animation scene to obtain scene picture elements;

[0051] A special effects rendering module, used to query the element type of the scene screen element, identify the degree of special effects demand in the element type, determine the special effects processing method corresponding to the initial animation screen based on the special effects demand degree, calculate the special effects processing index corresponding to the special effects processing method, and generate a special effects rendering solution corresponding to the animation scene based on the special effects processing index;

[0052] A color optimization module, used to query the color mode that the animation scene can adapt to, evaluate the visual atmosphere parameters of the animation scene, calculate the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and construct a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount;

[0053] A perspective adjustment module, for performing composition analysis on the screen layout of the animation scene to obtain composition analysis data, determining the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, querying the perspective parameters corresponding to the perspective type, and formulating a perspective adjustment plan for the animation scene based on the perspective parameters;

[0054] The intelligent rendering module is used to determine the rendering target corresponding to the initial animation picture based on the special effect rendering scheme, color optimization strategy and viewing angle adjustment plan, optimize the picture of the animation scene based on the rendering target, obtain a picture optimization record, monitor the rendering quality corresponding to the animation scene based on the picture optimization record, and generate an intelligent rendering scheme corresponding to the initial animation picture based on the rendering quality.

[0055] Firstly, by collecting initial animation pictures and identifying the animation scenes corresponding to the initial animation pictures, the present invention can quickly understand the overall style and atmosphere of the scenes, which is helpful for the rational allocation of resources. More computing resources can be allocated for rendering, while for simple indoor scenes, the resources can be appropriately reduced to avoid waste of resources, so that the resources can be perfectly integrated with the scenes, and the overall coordination and visual effects of the pictures can be improved. At the same time, by querying the element types of the scene picture elements and identifying the degree of special effects requirements in the element types, the present invention can enhance the narrative ability of the pictures, and design special effects according to the element types and the degree of special effects requirements. For example, when expressing the sadness of a character, adding appropriate blur effects to the tear elements can convey emotions more delicately and allow the audience to understand the plot more deeply. In addition, this detection and recognition can ensure the unity of the overall style of the animation and ensure that the special effects of each element are consistent with the scene style and plot atmosphere. The present invention can select the most appropriate color combination for the animation scene by querying the color mode that the animation scene can adapt to and evaluating the visual atmosphere parameters of the animation scene, thereby enhancing the beauty and emotional communication of the picture, ensuring that the color of the scene is coordinated with the overall style, and making the animation more visually unified. The present invention can determine the position, size and proportional relationship of each element in the picture by performing composition analysis on the picture layout of the animation scene to obtain composition analysis data, thereby creating a more balanced, harmonious and rhythmic picture, which can create different Atmosphere and emotion, the picture design and layout can be carried out more efficiently, avoiding unnecessary adjustments and modifications, and the placement and intensity of special effects, as well as the distribution and matching of colors can be determined to make the entire animation picture more coordinated and unified. Furthermore, the present invention determines the rendering target corresponding to the initial animation picture based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, which can add rich visual effects to the picture, such as light and shadow special effects, particle effects, etc., to make the scene more vivid and realistic; the color optimization strategy can adjust the color matching and tone of the picture to enhance the emotional atmosphere and artistic sense of the picture; the perspective adjustment plan can optimize the composition and perspective of the picture, thereby enhancing the overall sense and coherence of the work. Therefore, the intelligent rendering method and system for realizing animation pictures proposed by the present invention can improve the efficiency and quality of animation picture rendering. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of a flow chart of a method for realizing intelligent rendering of animation images provided by an embodiment of the present invention;

[0057] Figure 2 A schematic diagram of modules for implementing an intelligent rendering system for animation images provided by an embodiment of the present invention.

[0058] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0059] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0060] The embodiment of the present application provides a method for realizing intelligent rendering of animation pictures. The execution subject of the method for realizing intelligent rendering of animation pictures includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for realizing intelligent rendering of animation pictures can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0061] Embodiment 1:

[0062] Reference Figure 1 FIG. 1 is a flow chart of a method for realizing intelligent rendering of animation pictures provided by an embodiment of the present invention. In this embodiment, the method for realizing intelligent rendering of animation pictures includes:

[0063] S1. Collect an initial animation picture, identify the animation scene corresponding to the initial animation picture, analyze the animation effect requirements required for the animation scene, and based on the animation effect requirements and the corresponding style characteristics of the animation scene, perform picture analysis on the initial animation picture to obtain scene picture elements.

[0064] The present invention collects initial animation pictures and identifies the animation scenes corresponding to the initial animation pictures, so as to quickly understand the overall style and atmosphere of the scenes, contribute to the rational allocation of resources, and can allocate more computing resources for rendering. For simple indoor scenes, the resources can be appropriately reduced to avoid waste of resources, so that the resources can be perfectly integrated with the scenes, thereby improving the overall coordination and visual effects of the pictures.

[0065] Among them, the initial animation picture refers to the original image content formed in the early stage of animation production without further processing and optimization. It contains the initial presentation form of basic visual elements such as animation characters, objects, backgrounds, etc. The position, shape, color and other information of these elements are in the original setting state and are the starting material for a series of production processes such as subsequent rendering; the animation scene refers to the specific plot location in the animation with unique spatial and environmental characteristics. It covers various types, such as bustling urban street scenes, including elements such as building appearances, traffic facilities, street shops and bustling crowds on the streets; mysterious and fantastic magic forest scenes, with trees of various shapes, strange luminous plants, mysterious creatures hidden in the dark, etc.; or fiercely fighting battlefield scenes, with smoke, broken walls, and warring armies. Different animation scenes provide corresponding background environments for plot development and role performances. Optionally, the acquisition of the initial animation picture can be achieved through an image acquisition device, such as a Wacom digital tablet with drawing tools such as Photoshop, so as to accurately capture the brushstrokes, pressure and other information of the painting, and convert it into digital image data to form an initial animation picture; the identification of the animation scene corresponding to the initial animation picture can be achieved through a convolutional neural network model, such as ResNet, VGG and other models, and these models can be trained using a large amount of image data with annotated animation scene types. The trained model can analyze the image features in the initial animation picture, such as object shape, color distribution, texture and other information, and then identify the animation scene.

[0066] Furthermore, the present invention can ensure the best visual appeal of the picture by analyzing the animation effect requirements required for the cartoon scene. For example, for the scene of fierce battle, it can analyze the need for animation effects such as motion blur and flickering light and shadow. Secondly, it helps to improve the fluency of story narration and can reasonably arrange the animation effect requirements according to the importance of the scene, thereby ensuring high-quality presentation of the main scenes.

[0067] Among them, the animation effect requirements refer to a series of special dynamic presentation requirements determined according to the characteristics of the animation scene, the plot atmosphere and the artistic expression intention, which can enhance the visual expression and appeal of the scene, including but not limited to lighting-related effect requirements, such as strong light exposure, dynamic changes in light and shadow (such as the flickering and flow of light on the surface of objects), which can create different atmospheres such as mystery, sacredness or tension; dynamic change requirements in color, such as color gradients, conversion of warm and cold tones, which are used to set off the mood of the scene. For example, a sad scene may require a cool tone and a gradually dimming effect; there are also special material expression requirements, such as the reflective flickering effect of metal materials, the flow and refraction changes of water materials, to enhance the realism and texture of objects. Optionally, the animation effect requirements required for analyzing the animation scene can be achieved by an algorithm based on emotional semantic analysis, such as: TextBlob, VADER and other algorithms.

[0068] Furthermore, the present invention analyzes the initial animation picture based on the animation effect requirements and the corresponding style characteristics of the animation scene to obtain scene picture elements, which can accurately extract the core elements of the picture and make the picture more compact and expressive. For example, for a steampunk-style scene, the picture is analyzed in combination with the required animation effect requirements such as metal texture and gear rotation, and key elements such as steam pipes and mechanical parts can be accurately found, thereby avoiding cluttered picture elements and enhancing the visual hierarchy.

[0069] The style characteristics refer to the unique artistic style attributes presented by the animation scene, which is an overall style shaped by multiple elements, including the cultural background of the scene (such as oriental cultural style and western cultural style. For example, oriental style has Chinese ancient buildings and Japanese style elements, while western style can have European castles and Roman columns, etc.), era characteristics (such as ancient, modern, and future. Ancient style has simple buildings and traditional costumes, modern style has high-rise buildings and modern transportation, and future style has high-tech aircraft and advanced mechanical devices), visual expression techniques (such as realistic style, cartoon style, and abstract style. Realistic style focuses on the real shape and Texture, cartoon style has exaggerated shapes and bright colors, and abstract style uses a unique combination of geometric shapes and colors to convey artistic conception) and other aspects; the scene screen elements refer to the basic visual units that constitute the entire scene screen, which are extracted after a comprehensive analysis of the initial animation screen based on the animation effect requirements and scene style characteristics. For example, the scene screen elements include characters with specific appearances (hairstyles, clothing, etc.) and action postures, stylized buildings and natural objects (such as castles, trees, etc.), background elements with different dynamic effects (floating clouds, flowing water, etc.), and special effects that match the plot and style (such as magic light, battle smoke, etc.).

[0070] As an embodiment of the present invention, the initial animation picture is analyzed based on the animation effect requirement and the style characteristics corresponding to the animation scene to obtain scene picture elements, including: deconstructing the animation effect requirement to obtain a deconstruction feature set; identifying visual labels in the style characteristics corresponding to the animation scene; constructing a mapping rule corresponding to the initial animation picture based on the deconstruction feature set and the visual labels; based on the mapping rule, dividing the initial animation picture into regions to obtain divided picture regions; extracting regional visual points in the divided picture regions; based on the regional visual points, analyzing the initial animation picture to obtain scene picture elements.

[0071] The deconstructed feature set refers to a series of feature sets obtained after decomposing and analyzing the animation effect requirements. These features cover various attributes of the animation screen in terms of dynamic performance, including but not limited to action-related features (such as the speed, amplitude, and type of action, such as fast running, large-scale sword swinging, and different types of actions such as jumping and flying), light and shadow-related features (such as light intensity, direction, and color, such as direct strong light, backlight, warm light, etc.), color-related features (such as color type, color change mode, such as red as the main color, color The visual label refers to the logo of representative and iconic visual elements or features extracted from the style characteristics of the animation scene. For example, for a medieval European fantasy style scene, its visual labels may include knight's armor (with specific shapes and textures), magic runes (unique patterns), ancient castles (architectural features such as spires and thick walls), flying dragons (with appearance features such as wings and scales), etc. For a modern urban style scene, the visual label may be Skyscrapers (glass curtain walls, geometric shapes), neon lights (colorful, flashing lighting effects), cars (streamlined design), etc.; the mapping rule refers to a logical correspondence and guiding principle established between the deconstructed animation effect requirement feature set and the visual label of the animation scene style characteristics, which stipulates how to achieve the corresponding animation effect characteristics according to specific style labels, and is a bridge for organically combining animation effects with scene styles; the screen division area refers to the parts with different functions and visual meanings formed by dividing the initial animation screen according to the constructed mapping rules. For example, it can be divided into the character main area (this is where the main activities and displays of the story characters are located, including the animation effects such as the character's actions and expressions), the background area (including the large background of the scene, such as the sky, mountains, city views, etc., providing an environmental background for the characters and stories, and its animation effects may involve weather changes, the overall atmosphere of light and shadow, etc.), and the special effects area (an area specifically used to display special effects such as magic, smoke, and explosions; the regional visual point refers to an element or a collection of elements in the screen division area that has significant visual characteristics, can attract the audience's attention, or plays a key role in the style and animation effects of the area.

[0072] Furthermore, the deconstruction of the animation effect requirements can be achieved through a rule-based classification method, such as: through this hierarchical rule classification, the animation effect requirements are gradually deconstructed; the identification of visual labels in the corresponding style characteristics of the animation scene can be achieved through an image feature extraction algorithm, such as: SIFT, SURF and other algorithms; the construction of the mapping rules corresponding to the initial animation picture can be achieved through an association rule algorithm, such as: Apriori algorithm; the extraction of regional visual points in the divided areas of the picture can be achieved through a visual attention model, such as: inputting the image of the divided areas of the picture into the model, and the model outputs the most attention-grabbing point or area in each area, that is, the regional visual point; the picture analysis of the initial animation picture can be achieved through an image semantic segmentation algorithm, such as: Mask R-CNN and other algorithms.

[0073] S2. Query the element type of the scene screen element, identify the degree of special effect requirement in the element type, determine the special effect processing method corresponding to the initial animation screen based on the special effect requirement degree, calculate the special effect processing index corresponding to the special effect processing method, and generate a special effect rendering solution corresponding to the animation scene based on the special effect processing index.

[0074] The present invention can enhance the narrative ability of the picture by querying the element type of the scene screen element and identifying the degree of special effect demand in the element type, and designing special effects according to the element type and the degree of special effect demand. For example, when expressing the sad emotions of the character, adding appropriate blur special effects to the tear element can convey emotions more delicately and allow the audience to understand the plot more deeply. In addition, this detection and identification can ensure the unity of the overall style of the animation and ensure that the special effects of each element are consistent with the scene style and plot atmosphere.

[0075] Among them, the element type refers to the classification of each component in the animation scene based on its nature, function and visual characteristics, including character elements, such as protagonists and supporting roles, which have their own appearance, clothing, and action characteristics; object elements, such as weapons, props, and vehicles, which have different shapes, materials and uses; background elements, such as the sky, land, ocean, urban buildings, natural landscapes, etc.; the special effect requirement degree refers to the level or level of special effects required to be added to each element type in the picture in order to achieve ideal visual effects and artistic expression. For some key elements, such as the protagonist's weapons in the battle scene or the mysterious magic props that determine the direction of the plot, their special effect requirements are often high, and complex and cool special effects are required to highlight their status and functions, such as light and shadow trailing when weapons attack, and gorgeous light and energy fluctuation effects when magic props are activated. Optionally, the query of the element type of the scene screen element can be achieved through a semantic segmentation algorithm, such as: the scene screen can be divided into regions with different semantics, each region corresponding to an element type. By analyzing the segmentation results, the types of each element in the picture are determined; the identification of the degree of special effect demand in the element type can be achieved through classification algorithms in machine learning, such as support vector machine (SVM), decision tree and other algorithms.

[0076] Furthermore, the present invention determines the special effects processing method corresponding to the initial animation picture based on the degree of special effects demand, which can ensure the accurate delivery of special effects resources. For elements with high special effects demand, such as the core magic circle that releases powerful magic in a fantasy scene, complex and sophisticated special effects processing methods can be used, such as increasing the sense of layering of light, gorgeous changes in color, and dynamic texture effects, so that the magic circle becomes the focus of the picture and attracts the audience's attention. For elements with low special effects demand, such as ordinary buildings in the distance, simple light and shadow processing is used to present a harmonious and consistent visual effect.

[0077] Among them, the special effects processing method refers to a series of specific operating methods and processes used in the animation production process to achieve specific special effects. For example, for a character element that needs to add a magic attack special effect, the special effects processing method may include creating animation layers according to key special effects parameters in a specific software environment (such as Adobe After Effects, etc.), and adjusting the color, transparency and other visual parameters of the layer to achieve the color and layering of the magic light, and using keyframe animation to set the dynamic parameters such as the movement trajectory and speed of the light.

[0078] As an embodiment of the present invention, determining the special effects processing method corresponding to the initial animation picture based on the special effects requirement degree includes: quantitatively grading the special effects requirement degree to obtain a special effects requirement level; querying the special effects type template corresponding to the special effects requirement level; analyzing the style attributes of the elements in the initial animation picture to obtain style feature information; fusing the special effects type template with the style feature information to obtain a special effects fusion template; extracting key special effects parameters in the special effects fusion template; and determining the special effects processing method corresponding to the initial animation picture based on the key special effects parameters.

[0079] The special effect requirement level refers to the different levels of quantification of the special effect requirement, for example, it can be divided into three levels: low, medium and high, or more finely divided into levels one to five, etc. The low level may mean that only simple and basic special effects are needed, such as slight changes in light and shadow or simple dynamic effects, which are suitable for those elements that are not very noticeable in the picture or have little impact on the plot; the high level means that complex, delicate and visually impactful special effects are required, such as the gorgeous light and shadow of large magic arrays, skill collision special effects in fierce battle scenes, etc.; the special effect type template refers to the special effects template for A collection of pre-designed special effect modes for different special effect requirement levels. For example, for low-level special effect requirements, special effect type templates may include simple color gradients, single-directional light projection, etc.; for medium-level special effect requirements, templates can cover small explosion effects, local light and shadow flickering, etc.; and high-level special effect type templates have complex magic energy flow effects, multi-element combination special effect scenes (such as flames and lightning interweaving), etc.; the style feature information refers to the attribute set extracted from each element in the initial animation picture that can reflect the artistic style of the entire picture, which includes cultural style (such as ancient elements in oriental cultural styles, such as Chinese architecture and traditional clothing patterns, or European architecture and knight armor shapes in western cultural styles, etc.), era styles (ancient, modern or future styles, ancient styles have simple materials and decorations, modern styles have simple lines and technological elements, and future styles have high-tech and avant-garde designs), visual styles (realistic styles focus on the performance of real objects, cartoon styles have exaggerated shapes and colors, and abstract styles use unique geometric shapes and color combinations to convey emotions), etc.; the special effect fusion template refers to the special effect type template combined with style features. A new special effect design template generated after combining the special effect type template and the style feature information, for example, if the special effect type template is a flame special effect, and the style feature information is Japanese style, then in the special effect fusion template, the color of the flame may tend to be traditional Japanese red and orange tones, and the shape of the flame will have a more oriental painting style of fluidity and simplicity; the key special effect parameters refer to parameter information extracted from the special effect fusion template that plays a decisive role in the final presentation effect of the special effect, including the visual parameters of the special effect (such as color value, transparency, brightness, contrast, etc., which determine the appearance color and visual hierarchy of the special effect).

[0080] Furthermore, the quantitative grading of the degree of special effect demand can be achieved through machine learning classification algorithms, such as decision trees, random forests and other algorithms; the query of the special effect type template corresponding to the special effect demand level can be achieved through a hash table lookup method, such as creating a hash table, using the special effect demand level as a key and the corresponding special effect type template as a value, and by performing a hash operation on the special effect demand level, the corresponding special effect type template can be quickly located; the analysis of the style attributes of the elements in the initial animation picture can be achieved through an image style recognition algorithm, such as using a trained style recognition model, inputting the initial animation picture into the model, and outputting the style attributes of the picture by analyzing the texture, color distribution, object shape and other features in the picture; the fusion processing of the special effect type template with the style feature information can be achieved through a generative adversarial network, such as training a GAN Model, wherein the input of the generator is the special effect type template and style feature information, and the output is the fused special effect fusion template; the extraction of key special effect parameters in the special effect fusion template can be achieved through a model-based parameter extraction method, such as: in the 3D modeling software Maya, its API can be used to obtain the material properties (color, transparency and other parameters), animation curves (speed, motion trajectory and other parameters) and space transformation parameters (size, position and other parameters) of the special effect model; the determination of the special effect processing method corresponding to the initial animation screen can be achieved through special effect processing tools, such as: Adobe After Effects, Blender and other tools.

[0081] The present invention can clearly know the pros and cons of a certain special effect processing method in visual presentation by calculating the special effect processing index corresponding to the special effect processing method. For example, a high index means that the processing method can bring more shocking and more expected special effects, while a low index indicates that improvement may be needed. Secondly, it can be analyzed which special effect processing methods consume less resources while achieving a certain visual effect, avoiding excessive investment in inefficient special effect processing, and improving production efficiency.

[0082] Among them, the special effects processing index refers to a quantitative indicator used to measure the comprehensive performance of a specific special effects processing method in an animation scene. It comprehensively considers the quantitative value of the effect of key special effects parameters at different time points, the degree of fit with the overall style of the animation, and the audience's attention value, while taking into account the synergy effect of special effects processing methods in different dimensions and resource consumption.

[0083] As an embodiment of the present invention, the calculating of the special effect processing index corresponding to the special effect processing mode includes:

[0084] The special effect processing index corresponding to the special effect processing method is calculated using the following formula:

[0085] ;

[0086] in, Indicates the special effect processing index corresponding to the special effect processing method, Indicates the number of categories of key special effect parameters, Indicates the category index of the key special effect parameters. and Respectively represent the start time and end time of the special effect processing method corresponding to the animation timeline, Indicates at time At that time, The effect quantization value corresponding to the key special effect parameters of the class, Indicates at time At that time, The degree of compatibility between the key special effect parameters and the overall style of the animation. Indicates at time At that time, The attention value corresponding to the key special effect parameters of the class, represents the normalization constant, represents the synergistic effect matrix of the special effect processing methods in different dimensions, Indicates the number of types of occupied resources, Indicates the type index of occupied resources. Indicates The resource consumption of each type of occupied resources.

[0087] In detail, the key special effect parameters refer to the classification of various factors that have an important impact on the performance of animation special effects. Each type of key special effect parameters has its unique role. For example, color parameters affect the visual appeal and emotional expression of special effects, and speed parameters can determine the dynamics and tension of special effects. The animation timeline refers to a linear structure used to represent the time sequence of animation playback in animation production, covering the time periods when various special effects appear and change in the animation. The effect quantification value refers to a numerical value that quantifies the visual effect produced by the "n" type of key special effect parameters at a specific time "t". For example, for color parameters, it can be quantified according to factors such as color vividness, contrast, and harmony; for light and shadow parameters, it can be quantified according to light intensity, clarity of shadows, etc.; the fit refers to the degree of conformity between the "n" type of key special effect parameters and the overall style of the animation at a specific time "t"; the attention value refers to the value indicating that the "n" type of key special effect parameters at a specific time "t" are consistent with each other. The normalization constant refers to the value used to normalize the integration results of different key special effect parameters within a time range; the synergy effect matrix refers to a matrix used to measure the synergy of special effect processing methods in different dimensions. For example, the synergy of color and light and shadow may produce a stronger visual effect, and the synergy of speed and size may affect the dynamic sense of special effects; the occupied resources refer to the various types of resources consumed in the process of implementing special effect processing methods, including computing resources (such as CPU occupancy, memory usage), storage resources (such as hard disk space occupancy), rendering resources (such as rendering time, graphics processing unit usage), etc.; the resource consumption refers to the quantitative value of the specific consumption degree of the "m" type of occupied resources. For example, if it is computing resources, the resource consumption can be the percentage of CPU occupancy or the size of memory usage; if it is rendering resources, the resource consumption can be the length of time spent on rendering a specific special effect.

[0088] The present invention generates a special effect rendering scheme corresponding to the animation scene based on the special effect processing index, and can accurately understand which scheme is most suitable for the current animation scene. The special effect rendering scheme thus generated can meet the scene requirements to the greatest extent, and can provide a reference for subsequent animation production, thereby improving the quality and level of animation production.

[0089] Among them, the special effects rendering scheme refers to a set of detailed special effects production and presentation plans generated for a specific animation scene based on an evaluation of the special effects processing method. For example, for a fantasy battle scene, the special effects rendering scheme may include deciding at which key moments the gorgeous magic light effects appear, and how to set the color, intensity, dynamic changes and other parameters of these magic lights; determining the special effects expression form when the character's skills are released in battle, such as the shape, size, burning speed, etc. of the flame special effects. Optionally, the special effects rendering scheme corresponding to the animation scene generated based on the special effects processing index can be implemented through special effects preview tools, such as Houdini, Unreal Engine and other tools.

[0090] S3. Query the color mode that the animation scene can adapt to, evaluate the visual atmosphere parameters of the animation scene, calculate the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and construct a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount.

[0091] By querying the color mode that the animation scene can adapt to and evaluating the visual atmosphere parameters of the animation scene, the present invention can select the most appropriate color combination for the animation scene, enhance the beauty and emotional communication of the picture, ensure that the color of the scene is coordinated with the overall style, and make the animation more visually unified.

[0092] Among them, the color mode refers to the specific color combination and color adjustment style used in the animation scene, which includes the matching of color attributes such as hue, saturation, and brightness, as well as the proportional relationship between different colors. For example, the warm color mode may be dominated by warm colors such as red, orange, and yellow, creating a lively and warm atmosphere; the cold color mode is dominated by cold colors such as blue, green, and purple, conveying a sense of tranquility and mystery; the visual atmosphere parameter refers to the overall visual experience and emotional tone created by the animation scene through various visual factors such as picture elements, colors, light and shadow. For example, a dark and gloomy scene may have low brightness, high contrast, and cold-toned visual atmosphere parameters, giving people a feeling of tension and terror; while a bright and cheerful scene may have high brightness, low contrast, and warm-toned visual atmosphere parameters, which can make people feel relaxed and happy. Optionally, the query of the color mode that the animation scene can adapt to can be achieved through image analysis tools, such as GIMP, Adobe Photoshop, etc.; the evaluation of the visual atmosphere parameters of the animation scene can be achieved through data analysis tools, such as eye tracking equipment, heat map analysis tools, etc.

[0093] Furthermore, the present invention calculates the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and can make detailed adjustments to the colors according to specific visual atmosphere parameter requirements to make the picture more vivid and attractive, and can determine the color adjustment plan more quickly, reducing the time and cost of repeated attempts.

[0094] The color adjustment amount refers to the quantitative value of the degree of color change required to match the color of the animation scene with a specific visual atmosphere, which can be the adjustment value of the color brightness, contrast, saturation, hue, etc.

[0095] As an embodiment of the present invention, the step of calculating the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameter includes:

[0096] The color adjustment amount corresponding to the animation scene is calculated using the following formula:

[0097] ;

[0098] in, represents the color adjustment amount corresponding to the animation scene, represents the number of factors influencing the visual atmosphere parameter, Indicates the quantitative index of influencing factors, Indicates The visual atmosphere value corresponding to each influencing factor is: Indicates The weight coefficient of each influencing factor is Indicates The difficulty adjustment coefficient of each influencing factor, represents the number of color channels in the animation scene, Indicates the number index of color channels, Indicates The base color value for each color channel.

[0099] In detail, the influencing factors refer to various factors that can affect the visual atmosphere of the animation scene, which may include but are not limited to brightness, contrast, hue, color saturation, object shape, picture composition, etc.; the visual atmosphere value refers to the quantitative evaluation value of the visual atmosphere created by the "p"th influencing factor in the current animation scene. For example, if the influencing factor is brightness, then the visual atmosphere value can be the degree of difference between the brightness level of the current scene and the brightness level required for the ideal visual atmosphere; the weight coefficient refers to the importance weight assigned to the "p"th influencing factor. For example, if in a specific animation scene, the influence of hue on the visual atmosphere is greater than that of brightness, then the weight coefficient of the hue factor will be relatively high; the difficulty adjustment coefficient refers to the weight of the importance of the "p"th influencing factor. The difficulty of adjusting the influencing factors. Some factors are difficult to adjust due to technical limitations, scene complexity or other reasons. In this case, the difficulty adjustment coefficient will be higher; the color channel refers to the different channels used to represent color information in a digital image. Common color channels include red, green, and blue channels. Other channels may also be included as needed, such as a transparency channel; the basic color value refers to the initial color intensity value of the "v"th color channel before color adjustment is performed, which represents the initial state of the brightness, saturation or other color attributes of the color channel in the image.

[0100] The present invention constructs a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount, and can determine the color combination that best suits the animation theme and emotional expression, and then optimizes it in combination with the color adjustment amount, so that the color of the picture is richer, more harmonious, and more layered, thereby helping to enhance the artistic expression of the animation.

[0101] Among them, the color optimization strategy refers to a specific action plan and method formulated based on the color optimization direction, which is used to optimize the color of the initial animation screen. It may include selecting appropriate color adjustment tools and techniques, determining specific color adjustment steps and parameters, and formulating a color optimization schedule and resource allocation plan.

[0102] As an embodiment of the present invention, the color optimization strategy corresponding to the initial animation picture is constructed based on the color pattern and the color adjustment amount, including: extracting features of the color pattern to obtain pattern feature information; querying a color rule set that matches the pattern feature information; integrating the color adjustment amount with the color rule set to obtain integrated color data; analyzing the color optimization direction corresponding to the integrated color data; and constructing a color optimization strategy corresponding to the initial animation picture based on the color optimization direction.

[0103] Among them, the pattern feature information refers to a set of information extracted from a specific color mode that can represent the characteristics of the color mode, which may include characteristics such as the main color tone, the cold and warm tendency of the color tone, the saturation range of the color, the level of color contrast, etc. For example, in a color mode dominated by warm tones, the pattern feature information may include orange and yellow as the main tones, warm tones, high saturation, and moderate contrast; the color rule set refers to a set of color application rules formulated for different color mode characteristics. These rules may include color matching principles, color transition methods, color emphasis and weakening methods, etc. For example, for a color mode with high saturation and high contrast, the color rules The set includes rules such as boldly using complementary colors, adopting vivid color transitions to enhance visual impact, and highlighting the colors of main elements to attract attention; the integrated color data refers to a new data set obtained by fusing the color adjustment amount with the color rule set, and the integrated color data may include specific color information such as adjusted color values, color proportions, and color transition parameters; the color optimization direction refers to the specific direction and goal on how to further optimize the color of the animation picture based on the analysis of the integrated color data, which may include adjusting the brightness of the color to enhance the layering of the picture, optimizing color matching to highlight the theme, and adjusting the saturation and contrast of the color to create a specific emotional atmosphere.

[0104] Furthermore, the feature extraction of the color pattern can be implemented by a principal component analysis algorithm, such as: representing the color pattern as a set of color data vectors, extracting the main feature vectors by the principal component analysis algorithm, and these feature vectors represent the main features of the color pattern; the querying of the color rule set matching the pattern feature information can be implemented by a decision tree algorithm, such as: constructing a decision tree model, taking the pattern feature information as input, and determining the corresponding color rule set by branch judgment of the decision tree; the integration of the color adjustment amount and the color rule set can be implemented by a linear combination method, such as: linearly combining the color adjustment amount and the parameters in the color rule set to obtain the integrated color data; the analysis of the color optimization direction corresponding to the integrated color data can be implemented by a machine learning classification algorithm, such as: using a machine learning classification algorithm, taking the integrated color data as input, training a classification model, and predicting the color optimization direction; the construction of the color optimization strategy corresponding to the initial animation screen can be implemented by a strategy generation tool, such as: in the strategy generation tool, different task types and priorities can be set, and specific task steps and time arrangements can be determined according to the color optimization direction to generate a project plan for the color optimization strategy.

[0105] S4. Perform a composition analysis on the screen layout of the animation scene to obtain composition analysis data, determine the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, query the perspective parameters corresponding to the perspective type, and formulate a perspective adjustment plan for the animation scene based on the perspective parameters.

[0106] The present invention performs composition analysis on the picture layout of the animation scene to obtain composition analysis data, and can determine the position, size and proportional relationship of each element in the picture, so as to create a more balanced, harmonious and rhythmic picture, which can create different atmospheres and emotions, and can more efficiently design and layout the picture to avoid unnecessary adjustments and modifications. It can determine the placement and intensity of special effects, as well as the distribution and matching of colors, so that the entire animation picture is more coordinated and unified.

[0107] The composition analysis data refers to the quantitative and descriptive data results obtained after analyzing the layout of the animation scene. These data may include the position coordinates, size, proportional relationship, numerical description of layout features, etc. of the elements.

[0108] As an embodiment of the present invention, the composition analysis of the picture layout of the animation scene to obtain composition analysis data includes: performing image segmentation processing on the animation scene to obtain segmented image areas; identifying key composition elements in the segmented image areas; classifying and arranging the key composition elements to obtain classified composition elements; analyzing layout features corresponding to the classified composition elements; and based on the layout features, performing composition analysis on the picture layout of the animation scene to obtain composition analysis data.

[0109] Among them, the segmented image area refers to dividing the animation scene image into different parts or areas according to certain rules and methods. For example, an animation scene image can be divided into a sky area, a ground area, a building area, etc.; the key composition elements refer to image elements with important visual significance and composition effect in the segmented image area. These elements may include characters, objects, background elements, etc.; the classified composition elements refer to the results obtained after classifying and arranging the identified key composition elements according to certain standards. For example, the key composition elements can be divided into different categories such as characters, objects, background, etc., or into main elements, auxiliary elements, decorative elements, etc.; the layout features refer to the characteristics and laws of the picture layout analyzed from the classified composition elements. These features may include the distribution pattern, symmetry relationship, proportional relationship, primary and secondary relationship, etc. of the elements.

[0110] Furthermore, the image segmentation processing of the animation scene can be achieved through a threshold-based segmentation method, such as: dividing the image into different areas by setting a specific pixel value threshold; the identification of key composition elements in the segmented image area can be achieved through a target detection algorithm, such as: YOLO, Faster R-CNN and other algorithms; the classification and organization of the key composition elements can be achieved through a clustering algorithm, such as: clustering according to the color, shape, size and other characteristics of the elements, and classifying similar elements into one category; the analysis of the layout features corresponding to the classified composition elements can be achieved through an image analysis algorithm, such as: ImageJ, MATLAB and other algorithms; the composition analysis of the picture layout of the animation scene can be achieved according to traditional composition rules, such as: symmetrical composition, golden section composition, rule of thirds composition, etc.

[0111] The present invention determines the perspective type of the picture rendering corresponding to the initial animation picture based on the composition analysis data, which can better highlight the key elements of the picture, create a unique atmosphere and emotion, and help maintain the consistency of the animation style. In addition, the perspective type determined according to the composition analysis data can be consistent with the overall style of the animation, making the transition between different pictures more natural and smooth, and avoiding perspective confusion.

[0112] Among them, the picture rendering refers to the process of processing the two-dimensional or three-dimensional models, materials, lighting and other elements in the animation scene to generate the final visible image, including drawing the surface details of the model, applying material properties to determine the appearance of the object, setting lighting effects to simulate the light exposure in the real world and other operations; the perspective type refers to the different angles and methods of observing the scene in the animation picture, mainly including looking down, looking up, looking straight, close-up, panoramic and other types. Optionally, the determination of the perspective type of the initial animation picture corresponding to the picture rendering can be achieved through a machine learning algorithm, such as: using a machine learning algorithm to analyze a large number of existing animation works, learning the relationship between different picture contents and perspective types, and then inputting the characteristics of the initial animation picture into the trained model to let the model predict a suitable perspective type.

[0113] Furthermore, the present invention can maintain a unified visual style of the animation picture in different scenes by querying the perspective parameters corresponding to the perspective type. Instead, it can directly query the parameters suitable for a specific perspective type and quickly apply them to the picture rendering, which can better highlight the key elements of the picture and create the desired atmosphere and emotion.

[0114] Among them, the viewing angle parameters refer to specific values ​​and settings used to determine the specific viewing angle type of the animation picture, which include but are not limited to the height of the viewing angle (such as the height value of the overhead viewing angle), angle (such as the inclination angle of the oblique viewing angle), distance (such as the distance between the close-up viewing angle and the photographed object), field of view (determines the size of the scene that can be seen in the picture), etc. Optionally, the query of the viewing angle parameters corresponding to the viewing angle type can be achieved through neural network prediction, such as: using a trained neural network model, inputting a feature description of the viewing angle type, and allowing the model to predict the corresponding viewing angle parameters.

[0115] Furthermore, the present invention formulates a perspective adjustment plan for the animation scene based on the perspective parameters, and can adjust the perspective in a targeted manner to make the picture composition more reasonable and beautiful, highlight key elements, enhance the expressiveness and attractiveness of the picture, and ensure the coherence and consistency of the animation scene, thereby making the picture style more unified.

[0116] Among them, the perspective adjustment plan refers to a specific plan for systematic adjustment of the perspective of the animation scene formulated after comprehensive consideration of factors such as picture characteristics, adjustment requirements, preliminary adjustment directions and simulation adjustment parameters, including clear adjustment steps, parameter changes required for each step, and expected effects.

[0117] As an embodiment of the present invention, the perspective adjustment plan of the animation scene is formulated based on the perspective parameters, including: analyzing the picture features corresponding to the perspective parameters; determining the adjustment demand points corresponding to the picture features; based on the adjustment demand points, planning the preliminary adjustment direction corresponding to the animation scene; simulating the perspective adjustment of the preliminary adjustment direction to obtain simulated adjustment parameters; and formulating the perspective adjustment plan of the animation scene based on the simulated adjustment parameters.

[0118] Among them, the picture features refer to the visual characteristics of the animation scene determined by specific viewing angle parameters, including but not limited to the composition form of the picture (such as symmetrical composition, rule of thirds composition, etc.), the distribution state of elements (the position relationship of characters and objects in the picture), the color expression effect (the changes in the brightness and saturation of colors under different viewing angles), the strength of the sense of space (the hierarchical relationship between distant and near scenes), etc.; the adjustment demand point refers to the specific aspects of the animation scene that need to be improved or optimized based on the analysis of the picture features, which can be the needs determined for the purpose of enhancing the expressiveness of the picture, highlighting specific elements, improving visual balance, etc.; the preliminary adjustment direction refers to the general direction of adjusting the viewing angle of the animation scene determined based on the adjustment demand point, which can include raising or lowering the viewing angle, rotating the angle, expanding or reducing the field of view, etc.; the simulated adjustment parameters refer to the specific values ​​or settings obtained in the process of simulating the viewing angle adjustment of the preliminary adjustment direction, and these parameters can include specific viewing angle height change values, angle rotation degrees, size changes in the field of view, etc.

[0119] Furthermore, the analysis of the picture features corresponding to the viewing angle parameters can be achieved through image analysis tools, such as: using image analysis tools, the color distribution, contrast, brightness and other features of the image can be analyzed according to the viewing angle parameters, so as to determine the picture features; the determination of the adjustment requirements corresponding to the picture features can be achieved through animation evaluation tools, such as: according to the picture features, possible problems can be automatically detected and corresponding adjustment requirements can be proposed; the planning of the preliminary adjustment direction corresponding to the animation scene can be achieved by applying empirical rules, such as: for the situation where the picture is too crowded, the preliminary adjustment direction of expanding the field of view or adjusting the position of the object can be adopted; the simulation of the viewing angle adjustment of the preliminary adjustment direction can be achieved through animation production tools, such as: Toon Boom Harmony, Adobe Animate and other tools, to simulate the viewing angle adjustment of the preliminary adjustment direction; the formulation of the viewing angle adjustment plan for the animation scene can be achieved through project management tools, such as: Trello, Asana and other tools.

[0120] S5. Based on the special effects rendering scheme, color optimization strategy and viewing angle adjustment plan, determine the rendering target corresponding to the initial animation picture, optimize the picture of the animation scene based on the rendering target, obtain a picture optimization record, monitor the rendering quality corresponding to the animation scene based on the picture optimization record, and generate an intelligent rendering scheme corresponding to the initial animation picture based on the rendering quality.

[0121] Based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, the present invention determines the rendering target corresponding to the initial animation picture, which can add rich visual effects to the picture, such as light and shadow effects, particle effects, etc., to make the scene more vivid and realistic; the color optimization strategy can adjust the color matching and tone of the picture to enhance the emotional atmosphere and artistic sense of the picture; the perspective adjustment plan can optimize the composition and perspective of the picture, thereby enhancing the overall sense and coherence of the work.

[0122] Among them, the rendering target refers to the desired final picture effect determined during the animation production process based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, which includes but is not limited to specific requirements in terms of the visual style of the picture, color atmosphere, degree of special effects presentation, visual experience brought by the perspective, etc. Optionally, the determination of the rendering target corresponding to the initial animation picture can be achieved through a machine learning classification algorithm, such as: support vector machine, decision tree and other algorithms.

[0123] Furthermore, based on the rendering target, the present invention optimizes the picture of the animation scene, obtains the picture optimization record, and performs targeted optimization on each element of the picture, whether it is the details of the characters, the texture of the scene or the presentation of special effects, to achieve the ideal effect, making the picture more exquisite, delicate and layered, and enhancing the overall sense of the work.

[0124] Among them, the picture optimization record refers to the detailed documentation generated in the process of optimizing the animation scene, including the specific steps of optimization, such as the steps for adjusting the color of a character and the operating procedures for modifying the shape of an object; the tools and parameters used for optimization, such as the use of a color adjustment tool in the drawing software and its parameter settings; the comparison of the element states before and after optimization, such as the comparison between the blurriness of the character before optimization and the clarity after optimization; and the evaluation of the overall optimization effect of the entire animation scene.

[0125] As an embodiment of the present invention, the image optimization of the animation scene is performed based on the rendering target to obtain a image optimization record, including: analyzing the image element requirements corresponding to the rendering target; extracting scene elements in the animation scene based on the image element requirements; querying the element status corresponding to the scene elements; analyzing the state optimization points corresponding to the element status; and optimizing the animation scene based on the state optimization points to obtain a picture optimization record.

[0126] Among them, the picture element requirements refer to the specific specifications and conditions for the visual presentation of the animation scene determined according to the rendering target, which covers all dimensions of the picture, including color hue, saturation, contrast and other requirements. For example, if the rendering target is to create a mysterious atmosphere, dark tones and high contrast may be required; the layout and proportional relationship of the elements in the composition, such as the position and size ratio of the main elements; the scene elements refer to the various components that make up the animation scene, including characters, such as the appearance, clothing, expression and other details of the characters; objects, such as buildings, props and other physical elements; background elements, such as environmental elements such as the sky and mountains and rivers; as well as lighting effects, atmosphere elements, etc.; the element state refers to the visual state currently presented by the scene elements. Characteristics and attributes include the color condition of the elements, such as whether it is bright or cast; the integrity and accuracy of the shape, such as whether the character's appearance meets the design standards; the clarity of the elements, whether it is clear and sharp or blurred; the display level of the elements, whether it is in the foreground, middle ground or background, and the occlusion relationship with other elements, etc.; the state optimization point refers to the specific aspects of the scene elements that need to be improved and enhanced by comparing the element status and the requirements of the picture elements. If the picture elements require a high-definition character image, but the current element status shows that the character edges are blurred, then the clarity of the character is a state optimization point; if the picture elements require a warm background to create an atmosphere, and the existing background color is cold, color adjustment becomes a state optimization point.

[0127] Furthermore, the analysis of the requirements of the picture elements corresponding to the rendering target can be achieved through a text analysis method, such as: extracting the description of the picture elements related to the rendering target through text analysis technology; the extraction of scene elements in the animation scene can be achieved through an image segmentation algorithm, such as: a threshold-based segmentation algorithm, a region growing algorithm, etc.; the query of the element state corresponding to the scene element can be achieved through an image attribute measurement tool, such as: using an image attribute measurement tool to obtain the color information (RGB value, hue, saturation, brightness, etc.), size information, position information and other basic attributes of the scene element, so as to determine its element state; the analysis of the state optimization point corresponding to the element state can be achieved through a difference comparison algorithm, such as: comparing the element state data with the picture element requirements in terms of numerical or feature differences to determine the optimization point of color adjustment; the picture optimization of the animation scene can be achieved through a rendering engine, such as: a Maya renderer, a Cycles renderer, etc.

[0128] The present invention monitors the rendering quality corresponding to the animation scene based on the picture optimization record, and can accurately determine whether the rendered scene meets the initially set picture element requirements, thereby ensuring the accuracy of color, the presentation of elements, and the correct use of special effects and viewing angles, so that the quality of the entire animation picture is stable and high-quality.

[0129] Among them, the rendering quality refers to the quality of the overall visual effect presented by the animation scene after rendering, which covers multiple aspects of performance, including the clarity of the picture, that is, whether the image is sharp, whether the details are rich and discernible, and whether there is no blur or jagged phenomenon; the accuracy and coordination of colors, whether the colors accurately restore the design intent, and whether the color matching between the elements is harmonious, whether it can create the expected atmosphere and emotional effect. Optionally, the monitoring of the rendering quality corresponding to the animation scene can be achieved through image optimization tools, such as: ImageOptim, TinyPNG and other tools.

[0130] Furthermore, the present invention generates an intelligent rendering solution corresponding to the initial animation picture based on the rendering quality, which can quickly and accurately find out the problems and advantages of the current picture, and can optimize the problem areas in a targeted manner, avoiding the time-consuming step of repeatedly trying different rendering settings in the traditional production process, making the production process smoother and more efficient.

[0131] Among them, the intelligent rendering solution refers to a comprehensive and systematic rendering guidance plan generated by intelligent technology and algorithms based on the evaluation and analysis of the rendering quality of animation scenes. It covers multiple aspects, including but not limited to rendering strategies for color, such as automatically determining specific solutions for adjusting color balance, hue, saturation and other parameters based on the color deviation or incoordination of the picture; for picture clarity, according to the problems such as blur and jaggedness that appear in the rendering quality, plan appropriate resolution, anti-aliasing methods and sharpening measures; in terms of element presentation, for the shape, proportion and integrity problems of elements such as characters and objects found in the rendering quality detection, formulate accurate model rendering optimization solutions, such as improving material properties and lighting effects to make the elements more realistic and complete. Optionally, the intelligent rendering solution corresponding to the generation of the initial animation picture can be implemented through rendering simulation tools, such as: Blender, Maya and other tools.

[0132] Firstly, by collecting initial animation pictures and identifying the animation scenes corresponding to the initial animation pictures, the present invention can quickly understand the overall style and atmosphere of the scenes, which is helpful for the rational allocation of resources. More computing resources can be allocated for rendering, while for simple indoor scenes, the resources can be appropriately reduced to avoid waste of resources, so that the resources can be perfectly integrated with the scenes, and the overall coordination and visual effects of the pictures can be improved. At the same time, by querying the element types of the scene picture elements and identifying the degree of special effects requirements in the element types, the present invention can enhance the narrative ability of the pictures, and design special effects according to the element types and the degree of special effects requirements. For example, when expressing the sadness of a character, adding appropriate blur effects to the tear elements can convey emotions more delicately and allow the audience to understand the plot more deeply. In addition, this detection and recognition can ensure the unity of the overall style of the animation and ensure that the special effects of each element are consistent with the scene style and plot atmosphere. The present invention can select the most appropriate color combination for the animation scene by querying the color mode that the animation scene can adapt to and evaluating the visual atmosphere parameters of the animation scene, thereby enhancing the beauty and emotional communication of the picture, ensuring that the color of the scene is coordinated with the overall style, and making the animation more visually unified. The present invention can determine the position, size and proportional relationship of each element in the picture by performing composition analysis on the picture layout of the animation scene to obtain composition analysis data, thereby creating a more balanced, harmonious and rhythmic picture, which can create different Atmosphere and emotion, the picture design and layout can be carried out more efficiently, avoiding unnecessary adjustments and modifications, and the placement and intensity of special effects, as well as the distribution and matching of colors can be determined to make the entire animation picture more coordinated and unified. Furthermore, the present invention determines the rendering target corresponding to the initial animation picture based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, which can add rich visual effects to the picture, such as light and shadow special effects, particle effects, etc., to make the scene more vivid and realistic; the color optimization strategy can adjust the color matching and tone of the picture to enhance the emotional atmosphere and artistic sense of the picture; the perspective adjustment plan can optimize the composition and perspective of the picture, thereby enhancing the overall sense and coherence of the work. Therefore, the intelligent rendering method and system for realizing animation pictures proposed by the present invention can improve the efficiency and quality of animation picture rendering.

[0133] Embodiment 2:

[0134] like Figure 2 FIG. 1 is a schematic diagram of a module of an intelligent rendering system for realizing animation images provided by an embodiment of the present invention.

[0135] The intelligent rendering system 200 for realizing an animated picture of the present invention can be installed in an electronic device. According to the functions to be realized, the intelligent rendering system 200 for realizing an animated picture can include a picture analysis module 201, a special effect rendering module 202, a color optimization module 203, a viewing angle adjustment module 204 and an intelligent rendering module 205. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0136] In this embodiment, the functions of each module / unit are as follows:

[0137] The picture analysis module 201 is used to collect the initial animation picture, identify the animation scene corresponding to the initial animation picture, analyze the animation effect requirements required by the animation scene, and analyze the initial animation picture based on the animation effect requirements and the style characteristics corresponding to the animation scene to obtain scene picture elements;

[0138] The special effect rendering module 202 is used to query the element type of the scene screen element, identify the special effect requirement degree in the element type, determine the special effect processing method corresponding to the initial animation screen based on the special effect requirement degree, calculate the special effect processing index corresponding to the special effect processing method, and generate the special effect rendering scheme corresponding to the animation scene based on the special effect processing index;

[0139] The color optimization module 203 is used to query the color mode that the animation scene can adapt to, evaluate the visual atmosphere parameters of the animation scene, calculate the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and construct a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount;

[0140] A perspective adjustment module 204 is used to perform composition analysis on the screen layout of the animation scene to obtain composition analysis data, determine the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, query the perspective parameters corresponding to the perspective type, and formulate a perspective adjustment plan for the animation scene based on the perspective parameters;

[0141] The intelligent rendering module 205 is used to determine the rendering target corresponding to the initial animation picture based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, optimize the picture of the animation scene based on the rendering target, obtain a picture optimization record, monitor the rendering quality corresponding to the animation scene based on the picture optimization record, and generate an intelligent rendering scheme corresponding to the initial animation picture based on the rendering quality.

[0142] In detail, the modules described in the intelligent rendering system 200 for realizing animation pictures described in the embodiment of the present invention adopt the same technical means as the intelligent rendering method for realizing animation pictures described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0143] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for realizing intelligent rendering of animation pictures, characterized in that: The method comprises: Collecting an initial animation picture, identifying an animation scene corresponding to the initial animation picture, analyzing the animation effect requirements required for the animation scene, and performing picture analysis on the initial animation picture based on the animation effect requirements and the style characteristics corresponding to the animation scene to obtain scene picture elements; Querying the element type of the scene screen element, identifying the degree of special effect requirement in the element type, determining the special effect processing method corresponding to the initial animation screen based on the special effect requirement, calculating the special effect processing index corresponding to the special effect processing method, and generating the special effect rendering scheme corresponding to the animation scene based on the special effect processing index; Querying the color mode that the animation scene can adapt to, evaluating the visual atmosphere parameters of the animation scene, and calculating the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, wherein calculating the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters includes: The color adjustment amount corresponding to the animation scene is calculated using the following formula: ； in, represents the color adjustment amount corresponding to the animation scene, represents the number of factors influencing the visual atmosphere parameter, Indicates the quantitative index of influencing factors, Indicates The visual atmosphere value corresponding to each influencing factor is: Indicates The weight coefficient of each influencing factor is Indicates The difficulty adjustment coefficient of each influencing factor, represents the number of color channels in the animation scene, Indicates the number index of color channels, Indicates The basic color value of each color channel is used to construct a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount; Performing composition analysis on the screen layout of the animation scene to obtain composition analysis data, determining the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, querying the perspective parameters corresponding to the perspective type, and formulating the perspective adjustment plan of the animation scene based on the perspective parameters; Based on the special effects rendering scheme, color optimization strategy and perspective adjustment plan, determine the rendering target corresponding to the initial animation picture, based on the rendering target, optimize the picture of the animation scene to obtain a picture optimization record, based on the picture optimization record, monitor the rendering quality corresponding to the animation scene, and based on the rendering quality, generate an intelligent rendering scheme corresponding to the initial animation picture.

2. The method for realizing intelligent rendering of animation pictures according to claim 1, characterized in that: Based on the animation effect requirements and the corresponding style characteristics of the animation scene, the initial animation picture is analyzed to obtain scene picture elements, including: Deconstructing the animation effect requirement to obtain a deconstructed feature set; Identifying visual tags in the style characteristics corresponding to the animation scene; Based on the deconstructed feature set and the visual label, construct a mapping rule corresponding to the initial animation picture; Based on the mapping rule, the initial animation screen is divided into regions to obtain screen division regions; Extracting regional visual points in the divided areas of the picture; Based on the regional visual points, the initial animation picture is analyzed to obtain scene picture elements.

3. The intelligent rendering method for realizing animation pictures according to claim 1, characterized in that: The determining, based on the degree of special effect demand, a special effect processing method corresponding to the initial animation picture includes: Quantitatively grading the degree of special effect demand to obtain a special effect demand level; Query the special effect type template corresponding to the special effect requirement level; Analyzing the style attributes of the elements in the initial animation picture to obtain style feature information; Fusing the special effect type template with the style feature information to obtain a special effect fusion template; Extracting key special effect parameters in the special effect fusion template; Based on the key special effect parameters, a special effect processing method corresponding to the initial animation picture is determined.

4. The method for realizing intelligent rendering of animation pictures according to claim 1, characterized in that: The step of constructing a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount includes: Extracting features of the color pattern to obtain pattern feature information; Querying a color rule set that matches the pattern feature information; Integrating the color adjustment amount with the color rule set to obtain integrated color data; Analyzing a color optimization direction corresponding to the integrated color data; Based on the color optimization direction, a color optimization strategy corresponding to the initial animation picture is constructed.

5. The method for realizing intelligent rendering of animation pictures according to claim 1, characterized in that: The performing composition analysis on the picture layout of the animation scene to obtain composition analysis data includes: Performing image segmentation processing on the animation scene to obtain segmented image areas; identifying key compositional elements in the segmented image region; Classify and sort the key composition elements to obtain classified composition elements; Analyzing layout features corresponding to the classified composition elements; Based on the layout features, a composition analysis is performed on the picture layout of the animation scene to obtain composition analysis data.

6. The method for realizing intelligent rendering of animation pictures according to claim 1, characterized in that: The step of formulating a perspective adjustment plan for the animation scene based on the perspective parameter includes: Analyzing picture features corresponding to the viewing angle parameters; Determining adjustment demand points corresponding to the picture features; Based on the adjustment demand point, planning a preliminary adjustment direction corresponding to the animation scene; Performing a simulated viewing angle adjustment on the preliminary adjustment direction to obtain a simulated adjustment parameter; Based on the simulation adjustment parameters, a viewing angle adjustment plan for the animation scene is formulated.

7. The method for realizing intelligent rendering of animation pictures according to claim 1, characterized in that: The step of optimizing the animation scene based on the rendering target to obtain a picture optimization record includes: Analyze the screen element requirements corresponding to the rendering target; Based on the picture element requirements, extracting scene elements in the animation scene; Querying the element status corresponding to the scene element; Analyze the state optimization point corresponding to the element state; Based on the state optimization point, the animation scene is optimized to obtain a picture optimization record.

8. An intelligent rendering system for realizing animation pictures, characterized in that: A method for implementing an intelligent rendering method for an animation screen as claimed in any one of claims 1 to 7, the system comprising: A picture analysis module is used to collect an initial animation picture, identify an animation scene corresponding to the initial animation picture, analyze the animation effect requirements required by the animation scene, and perform picture analysis on the initial animation picture based on the animation effect requirements and the style characteristics corresponding to the animation scene to obtain scene picture elements; A special effects rendering module, used to query the element type of the scene screen element, identify the degree of special effects demand in the element type, determine the special effects processing method corresponding to the initial animation screen based on the special effects demand degree, calculate the special effects processing index corresponding to the special effects processing method, and generate a special effects rendering solution corresponding to the animation scene based on the special effects processing index; A color optimization module, used to query the color mode that the animation scene can adapt to, evaluate the visual atmosphere parameters of the animation scene, calculate the color adjustment amount corresponding to the animation scene based on the visual atmosphere parameters, and construct a color optimization strategy corresponding to the initial animation picture based on the color mode and the color adjustment amount; A perspective adjustment module, for performing composition analysis on the screen layout of the animation scene to obtain composition analysis data, determining the perspective type of the screen rendering corresponding to the initial animation screen based on the composition analysis data, querying the perspective parameters corresponding to the perspective type, and formulating a perspective adjustment plan for the animation scene based on the perspective parameters; The intelligent rendering module is used to determine the rendering target corresponding to the initial animation picture based on the special effect rendering scheme, color optimization strategy and viewing angle adjustment plan, optimize the picture of the animation scene based on the rendering target, obtain a picture optimization record, monitor the rendering quality corresponding to the animation scene based on the picture optimization record, and generate an intelligent rendering scheme corresponding to the initial animation picture based on the rendering quality.

Images