926 results about "Computer graphics" patented technology

The invention discloses a large-scene three-dimensional reconstruction method based on three-dimensional Gaussian sputtering, and relates to computer graphics. The method comprises the following steps: collecting a multi-view image set of a large scene; obtaining a scene sparse point cloud according to the multi-view image set; performing monocular depth estimation on the multi-view image by using a pre-trained depth prediction network to obtain monocular depth estimation priori; the method comprises the following steps of: performing global training on a scene by utilizing scene sparse point cloud and monocular depth estimation prior to obtain an initial three-dimensional Gaussian model, and performing space grid division on the initial three-dimensional Gaussian model to obtain a plurality of scene blocks with axis alignment bounding boxes; setting image view angle data of each scene block; performing deep supervised training on the Gaussian ellipsoids in the plurality of scene blocks by using a parallel GPU (Graphics Processing Unit); combining the trained scene blocks to obtain a final three-dimensional Gaussian model; in view of low geometric structure reconstruction precision caused by only depending on color information of a multi-view image in large-scene three-dimensional rendering, the method improves the reconstruction precision of large-scene rendering.

According to an aspect of the present invention, there is provided an electrical audio signal processing system and device, comprising: a computer graphics processing and selective visual display system with a screen; an eye tracking device; a processor; one or more computer memory devices; wherein the processor is arranged for operations comprising: measuring the user's eye movements to ascertain the specific word on which the user is fixated, by the eye tracking device; modifying the display at the user's current fixation point; applying a delay between the presentation of successive graphic elements based on the user's calculated rate to accommodate the user's required time; and presenting elements to the user at a rate based upon the user's required time.

The invention discloses a global nerve drawing method and system for a multivariate mixed representation scene, and belongs to the technical field of computer graphics, and the method comprises a scene expression module which is used for loading multivariate mixed scene data; the geometric module is used for generating a multi-channel geometric buffer based on the multivariate mixed scene data; the illumination module comprises at least one neural network sub-module and is used for carrying out illumination calculation based on multi-channel geometric buffering; the scheduling and synchronizing module is used for coordinating cross-module data streams through a rendering graph structure and executing resource merging and task scheduling between the geometric module and the illumination module based on a unified request structure; and the merging output module is used for fusing neural network illumination output and traditional rendering output to generate a final frame image so as to complete global neural rendering. According to the invention, multi-element scene data is integrated through modular design, efficient collaboration of traditional and neural rendering is realized by using a unified interface and a scheduling structure, and the image quality and the system flexibility are improved.

The invention relates to the technical field of computer graphics, in particular to a real-time three-dimensional graphic rendering optimization method and device based on deep learning, and the method comprises the steps: carrying out the feature extraction of the geometric, texture and illumination information of a graphic through a multi-modal deep neural network, and generating a high-dimensional feature vector; dynamically adjusting the resolution of the triangular mesh by adopting a graph neural network, and outputting an optimized geometric mesh structure; predicting global illumination of the graph by using a neural radiation field model, and generating an optimized illumination map through a variational auto-encoder; performing fusion analysis on the illumination map and the texture data by using a visual Transform model to generate an adaptive texture mapping scheme; and in combination with a rendering scheduling strategy of deep reinforcement learning and a super-resolution reconstruction method of a generative adversarial network, a rendering result is optimized, and a rendered image is post-processed. According to the method, multiple deep learning technologies are comprehensively utilized, and the quality and efficiency of real-time three-dimensional graph rendering are remarkably improved.

The invention discloses a complex material-oriented global full-frequency illumination neural rendering method and system, and belongs to the technical field of computer graphics, and the method comprises the steps: querying and extracting a primary frequency feature and a secondary frequency feature in a neural feature field according to the primary geometric buffer and the reflection geometric buffer of a rendering pixel; performing weighted combination on the primary frequency features and the secondary frequency features through a double-frequency fusion module to form full-frequency fusion features; inputting the full-frequency fusion feature and the initial geometric buffer into a decoder to obtain a pixel radiation brightness value; and optimizing the network by adopting a double-stage training strategy of low-frequency data and multi-frequency data, and then using the optimized network for neural drawing of global full-frequency illumination. The method can improve the accuracy and stability of global illumination simulation in a complex dynamic scene, and is suitable for modeling a multi-frequency global illumination effect in a dynamic three-dimensional scene.

The embodiment of the invention provides a scene surface reconstruction method and device, equipment and a medium, and relates to the technical field of computer graphics. The method comprises the following steps of: firstly, introducing an initialized three-dimensional Gaussian model, and determining attribute information and position information of two-dimensional Gaussian corresponding to three-dimensional anchor Gaussian by combining a multi-view image sequence of a target scene; secondly, utilizing the attribute information and the position information to generate a rendered image through a Gaussian splashing method; and then, according to the rendered image and an image with the same view angle as the rendered image in the multi-view image sequence, updating the initialized three-dimensional Gaussian model by minimizing a rendering loss function. And finally, depth information is determined according to the updated three-dimensional Gaussian model and the multi-view image sequence, and a surface model of the target scene is constructed through a three-dimensional reconstruction algorithm according to the depth information and the rendered image, so that the structure and details of the target scene can be reflected more accurately, the floating object phenomenon is avoided, and the accuracy of the target scene is improved. And the quality and the accuracy of scene reconstruction are obviously improved.

The invention discloses a multi-mode driven virtual digital human face animation generation method and system, and relates to the field of computer graphics, and the method comprises the steps: obtaining voice input and text input, and extracting voice features and text features; dynamically fusing the two features through an attention fusion model to generate facial expression and head posture control parameters, and dynamically adjusting contribution weights of a voice mode and a text mode to the control parameters by adopting a driving strategy of differentiation of facial upper expression and facial lower expression; and performing local deformation on the virtual digital human face image based on the control parameter to generate an initial animation frame, and performing refining processing on the initial animation by using a generative adversarial network to obtain a refined face animation. By means of the technical scheme, the natural and vivid virtual human face animation matched with the voice content and the text semantics can be generated.

The invention relates to the technical field of computer graphic interactive processing, in particular to a real-time interactive image generation system based on a multi-point touch canvas. The input acquisition unit is used for acquiring original touch data from an operating system and preprocessing the original touch data; the gesture recognition and analysis unit is used for performing high-level semantic behavior analysis on the contact data sequence processed by the preprocessing module; the interaction control and parameter mapping unit is used for receiving the semantic event output by the gesture recognition and analysis unit, analyzing the semantic event into an image control command and generating an executable command sequence; and the image generation unit generates interactive image content dynamically responded in real time based on an internal graph state management mechanism. By introducing the adaptive Kalman filtering and trajectory prediction auxiliary mechanism, the filtering intensity of the contact data can be dynamically adjusted, the efficient suppression of finger jitter and the consistent reconstruction of the contact ID are realized, and the input stability and data continuity under the multi-point touch operation are remarkably improved.

The invention belongs to the technical field of three-dimensional reconstruction in computer graphics, particularly relates to a sparse view outdoor scene three-dimensional reconstruction method based on multi-view stereo and grids, and aims to solve the problems that input depends on dense views, rendering results are prone to generating fuzzy artifacts, large-scale scene memory occupation is high and the like in the existing reconstruction technology. The method is especially suitable for high-precision reconstruction of an outdoor boundless scene under sparse view input. The method comprises the following implementation steps: acquiring a multi-view image of a scene to be reconstructed and a corresponding camera pose, and determining a scene center based on a K-means clustering algorithm; extracting input view features, generating a cost body based on a multi-view three-dimensional framework, and obtaining scene depth distribution priori by combining a probability depth estimation method; guiding voxel grid sampling based on depth distribution, and mapping light rays exceeding a grid boundary into a limited grid range by using a spatial contraction function; extracting image local features and voxel grid global features for adaptive fusion; and finally, generating a reconstructed image through volume rendering.

The invention discloses a point cloud model rendering optimization method and system based on cloud edge collaboration, and belongs to the field of computer graphic processing, and the method comprises the steps: carrying out the spatial partitioning of multi-stage point cloud data, generating a point cloud hierarchical structure which can be quickly accessed, building a user portrait model, and predicting the region of interest of the user portrait model; the multi-level LOD point cloud data structure is distributed to an edge computing node, the current network state is evaluated in real time, and if the local cache of the edge node cannot meet the rendering requirement, the edge node adaptively requests for cloud incremental data; the client receives the data blocks from the edge nodes and uploads real-time interaction data to the edge nodes; and the client interaction record and the edge node cache hit rate are transmitted back to the cloud end, and the cloud end performs iterative optimization on a point cloud data distribution strategy by using a reinforcement learning model. According to the method, priority loading and high-precision rendering of the region of interest are achieved by constructing the multi-level LOD point cloud data structure, the delay and jamming phenomena are effectively avoided, and the user interaction experience is improved.

The invention discloses a self-adaptive 4D Gaussian splashing high-precision three-dimensional reconstruction system and method, and belongs to the technical field of computer graphic processing. The invention aims to realize high-precision automatic registration of multi-source heterogeneous data and improve the calculation efficiency. The method comprises the following steps: collecting multi-source heterogeneous data; constructing a multi-modal fusion registration method, which comprises the following steps: combining satellite image data and low-altitude oblique photography data to realize spatial distribution geometric coarse registration, fusing low-altitude laser radar point cloud data and ground acquisition vehicle laser radar point cloud data to realize luminosity fine registration, establishing semantic features to assist registration, and obtaining registered multi-source data; initializing a 4D Gaussian primitive and executing adaptive splashing reconstruction to obtain an optimized 4D Gaussian splashing model; designing a cloud edge cooperative computing architecture oriented to 4D Gaussian splash reconstruction, and performing distributed parallel processing on the obtained optimized 4D Gaussian splash model; and executing quality evaluation and adaptive optimization, and outputting a final adaptive 4D Gaussian splash model.

A full-automatic movie and television video production system based on digital actors belongs to the field of artificial intelligence, computer graphics and movie and television production and comprises a creative engine unit, a production unit, an intelligent shooting unit, a post-processing unit and a quality inspection and rendering unit. The method has the advantages that 1, the manufacturing efficiency is greatly improved, the traditional film and television manufacturing period is usually 6-12 months, the manufacturing period can be shortened to 4 hours by the method, and the efficiency is improved by about 1000 times; 2, the manufacturing cost is obviously reduced: the manufacturing cost of traditional movies and other movies is usually 5 million to 20 million dollars, but the method only needs the computing power cost, and the cost is less than one percent of that of the traditional movies; 3, the style consistency is guaranteed, traditional film and television production depends on personal ability, the style consistency is difficult to guarantee, and the style consistency is strictly guaranteed through an algorithm and is improved by 100%.

The invention discloses a Gaussian splash model training method and device, a medium and a program product, and relates to the technical field of computer graphics, and the method comprises the steps: carrying out the initialization of an original Gaussian splash model based on a sparse input sample image; according to the sparse input sample image and a preset Gaussian densification strategy, multiple rounds of iterative training are carried out on the original Gaussian splash model to obtain a progressive Gaussian splash model, and the Gaussian densification strategy comprises the following steps: when the number of times of iterative training of the original Gaussian splash model meets a preset number range, the number of times of iterative training of the original Gaussian splash model meets the preset number range; the densification threshold value used by the original Gaussian splash model in the iterative training process is inversely proportional to the number of iterative training times. According to the method, a relatively high densification threshold value is used at the initial stage of training, so that an over-fitting risk caused by over-high densification degree at the initial stage of training under a sparse input view angle is reduced.

The invention belongs to the technical field of computer graphics and real-time 3D scene reconstruction, and relates to a real-time compression and optimization method based on a 3D Gaussian sputtering (3DGS) model, which realizes efficient storage, transmission and rendering of a three-dimensional scene through residual vector quantization (RVQ) and a dynamic pruning technology, and comprises the following steps: S1, residual quantization vector quantization; s2, enhancing a dynamic pruning strategy; and S3, index compression and coding optimization. The invention provides a real-time 3D Gaussian sputtering (3D Gaussian sputtering) model compression system combining residual vector quantization and dynamic pruning, which can realize high-quality real-time rendering and extremely high storage efficiency. Through three core technologies of a vector quantization strategy, dynamic pruning strategy enhancement and index compression and coding optimization, on the premise that the rendering quality is guaranteed, the purposes that the model storage amount is reduced by more than 45 times, and the rendering speed is increased by 3 times are achieved.

The invention relates to the technical field of computer graphics, and provides a rendering method and device for a static scene under a dynamic view angle, equipment and a medium, and the method comprises the steps: obtaining a future view angle probability cloud picture of a preset frame number according to the historical behavior data of a user and a hybrid neural network view angle prediction model; dividing a pre-rendering static scene of at least one target pre-rendering view angle based on a preset dynamic semantic weight matrix to obtain layered baking data; obtaining target optical flow data of a target adjacent view angle adjacent to the current view angle from a preset optical flow mapping table; during view angle switching, dynamically mixing a static scene rendered in real time at the current view angle and a static scene pre-rendered according to a pixel displacement relation based on target optical flow data to obtain scene mixed data; and rendering based on the scene mixed data to obtain a target static scene. The problem of real-time high-quality rendering of the static scene under the dynamic view angle can be solved.

The invention relates to the fields of computer graphic processing, industrial software technology and the like, discloses a Cocos engine-oriented pipeline loop perception linear decomposition method, and aims to solve the problem that a complex topological graph which cannot be effectively processed and comprises a high-density nested loop in the prior art. The structure is completely and accurately decomposed into a linear drawing instruction sequence which can be accepted by a Cocos engine, and lossless conversion of original topological information is ensured; comprising the steps that after undirected topological graph data are input, a ring detection algorithm is called to recognize and record all independent ring structures in an undirected topological graph and nodes and edges contained in the independent ring structures; judging whether at least one ring is detected in the previous step, and if the at least one ring exists, assigning a determined traversal direction for each detected ring according to a preset ring direction rule; and if the ring does not exist, skipping, and directly performing depth-first search traversal on the undirected topological graph data.

The invention provides a fractal graph drawing method and device, a storage medium and electronic equipment, and relates to the field of computer graphics. The method comprises the following steps: the electronic equipment iterates a preset number of times by utilizing a current basic fractal graph to obtain a derivative fractal graph; judging whether the number of pixels required for displaying the derivative fractal graph is changed or not compared with the current basic fractal graph; if the derivative fractal graph will change, taking the derivative fractal graph as a new basic fractal graph, and returning to the step of obtaining the derivative fractal graph by utilizing the current basic fractal graph to carry out iteration for preset times; and if the fractal graph does not change, taking the derivative fractal graph as a target fractal graph. Thus, whether the fractal graph reaches the optimal definition or not is judged by judging whether the number of pixels before and after iteration changes, and compared with visual inspection, the definition of the fractal graph can be judged more objectively.

The invention discloses an interactive garden design method and system based on landscape garden simulation, particularly relates to the technical field of garden landscape design and computer graphic simulation, and relates to a plurality of modules of environmental data acquisition, three-dimensional modeling, plant growth simulation, water simulation, interactive design, microclimate prediction and the like. According to the method, terrain, climate and soil data are collected through a laser radar and multiple sensors, and a standardized environmental map is constructed; generating a plant topological structure based on a context sensitive type L system, and simulating a growth process by adopting a Logistic function in combination with an environmental factor; the method comprises the following steps: realizing dynamic simulation of a water body by utilizing an SPH particle method and a Navier-Stokes equation, and establishing a plant-water body feedback model; and real-time operation and design optimization are supported through a three-dimensional interaction interface. The system has the advantages of being high in ecological adaptability, high in visualization degree, perfect in feedback mechanism and the like, the intelligence and scientificity of garden design are remarkably improved, and the system is suitable for various scenes such as urban green lands and ecological restoration.

The invention provides a cuttable interactive software simulation method and system based on a GPU accelerated substance point method, and belongs to the field of computer graphics physical simulation, and the method comprises the steps: S1, carrying out the sampling and initialization of a to-be-simulated three-dimensional software model and a rigid body model interacting with the to-be-simulated three-dimensional software model; s2, based on a virtual scene to be simulated, constructing an initial GPU secondary sparse acceleration structure Ggrid; s3, carrying out expansion on the Ggrid; s4, updating the state of the rigid body model, and constructing a rigid body cutting field of Ggrid; s5, on the basis of the Ggrid and the rigid body cutting field, the physical explicit integral of the three-dimensional software model to be simulated is solved in parallel, and the position of the lagrangian particle at the next moment is obtained; and S6, during the last iteration of each frame, reconstructing a surface triangular mesh of the Lagrange particle cluster in real time based on Ggrid, and smoothing the surface triangular mesh. According to the method disclosed by the invention, multi-material soft deformation, cutting interaction, crushing and real-time surface reconstruction are realized.

The invention discloses a multi-level detail automatic simplification method for an oblique photography live-action three-dimensional model, and relates to the technical field of three-dimensional model simplification and computer graphics, and the method comprises the steps: obtaining oblique photography original data and three-dimensional model basic information; preprocessing the model, performing adaptive Gaussian filtering denoising, improving RANSAC to remove outer points, compressing textures in a blocking manner, correcting mapping coordinates, and repairing a topological structure; extracting multi-scale features; constructing a simplified decision model, and determining a simplification rate and a priority by combining an observation distance, scene precision and hardware performance; performing hierarchical simplification, vertex hierarchical improved edge folding, patch hierarchical adaptive deletion and regional hierarchical grid reconstruction; performing multi-dimensional quality evaluation, and if the requirements are not met, performing backtracking adjustment; and outputting a simplified model stored according to the LOD hierarchy, wherein the simplified model comprises transition information and a simplified log. According to the method, the data quality is improved through refined preprocessing, the simplification pertinence is enhanced through multi-dimensional feature extraction and intelligent decision, and the application value of the model is improved.

The invention discloses a WebGL-based mass data particle system rendering method and device, and belongs to the technical field of computer graphic processing. The method comprises the following steps: firstly, selecting the type of a particle emitter according to scene requirements, then constructing a primitive object based on points or quadrangles, and configuring three-dimensional coordinates and texture parameters; aggregation rendering is carried out on a single-frame primitive array through a WebGL vertex shader and a fragment shader, and the drawing calling frequency is reduced; the method comprises the steps of dynamically adjusting particle attributes, including position compensation, color gradient and explicit-implicit control, determining a rendering form and playing time of an animation according to requirements of an actual application scene, and supporting fine configuration of a single particle, including texture animation, dynamic scaling and rotation parameters. When a large amount of data is processed, on the premise of ensuring the rendering quality, the rendering efficiency can be remarkably improved, the consumption of a memory and GPU resources can be reduced, the response time of the system can be optimized, and meanwhile, the interaction experience of a user can be improved.

The invention discloses a general nerve drawing method and system based on an illumination function generation model, and belongs to the technical field of computer graphics, and the method comprises the steps: collecting light source information containing multi-view observation data, and collecting scene information; constructing an illumination function generation model comprising a light source coding module and a light source decoding module for converting the light source information into neural illumination representation and performing joint inference based on the neural illumination representation and scene features of the drawing points; and generating a final drawn image conforming to the real illumination distribution based on the inference result. According to the method, generalization neural drawing across light sources and scenes can be achieved, images with off-line rendering quality can be generated at the cost close to real-time calculation, the sense of reality, stability and rendering efficiency are considered, and the method has good expansibility and wide application value.

The invention belongs to the technical field of computer graphics and virtual reality, and discloses a virtual-real fusion-based panoramic roaming construction method for a digital exhibition hall, which effectively solves the problems of easy drift and offset in a virtual environment in a traditional system through a multi-sensor fusion algorithm and a closed-loop optimization mechanism. Under the complex environment of people flow change, illumination fluctuation and the like in an exhibition hall, accurate alignment of virtual content and physical space can still be kept, it is ensured that a user obtains coherent and stable visual experience in the roaming process, and a reliable space reference is provided for virtual-real interaction; the light and shadow effect and the material performance of the virtual content are naturally fused with the physical environment through the physical rendering and multi-sensory interaction technology, and meanwhile, a multi-dimensional immersion sense is constructed in combination with the spatial sound effect and interaction feedback; when the user interacts with the virtual exhibit, sensory experience close to reality can be obtained, and the sense of substitution and attraction of the digital exhibition hall are remarkably improved.

The invention discloses a multi-target tracking method, system and device, a medium and a product, and relates to the technical field of computer graphic vision, and the method comprises the steps: based on a plurality of single-view two-dimensional images, employing a lightweight key point detection model to extract human body two-dimensional key point coordinates under different views; based on the camera parameters corresponding to different visual angles and the human body two-dimensional key point coordinates under different visual angles, performing three-dimensional reconstruction on the human body key points by using a triangulation method to obtain a three-dimensional reconstruction model; projecting the three-dimensional reconstruction model, and determining a projection frame corresponding to each target in the tracked area; and a tracking algorithm and Kalman filtering are combined to complete real-time tracking of a plurality of targets in the tracked area. According to the method, the three-dimensional target tracking problem is reduced into the two-dimensional target tracking problem based on the projection frame of the three-dimensional reconstruction model relative to the horizontal plane, the accuracy and robustness of the model can be improved in a multi-target tracking scene, and meanwhile, the calculation amount is reduced.

The invention provides a training-free text-image generation method based on a diffusion model, and relates to the technical field of computer graphic processing and artificial intelligence. The method comprises the following steps: extracting semantic phrases and layout information in an input text by utilizing a natural language model, inputting the input text, the semantic phrases and the layout information as additional conditions into a diffusion model, and extracting cross attention maps of different time steps; a positive and negative sample concept and a foreground and background concept based on an object are constructed, a new loss function is calculated on a cross attention map for semantic information and layout information, the loss function combines semantic loss, regional loss and original loss of a diffusion model and is used for updating a potential space image, and the image is generated through iterative denoising and a decoder. According to the method, additional training is not needed, image generation output based on the diffusion model better meets text requirements, and a better text and image alignment effect is achieved.

View synthesis is a computer graphics process that generates a new image of a scene from a novel (previously unseen) viewpoint of the scene. Typically, the graphics process relies on a machine learning model that has been trained with ground truth pose information. Since ground truth pose information is not readily available, some solutions rely on a Structure-from-Motion (SfM) library COLMAP to generate pose information for a given image. However, this pre-processing step is not only time-consuming but also can fail due to its sensitivity to feature extraction errors and difficulties in handling texture-less or repetitive regions. The present disclosure provides view synthesis from learned camera poses without relying on SfM pre-processing.

The invention discloses a graphic processing engine development method and system based on virtual simulation, and relates to the technical field of computer graphics and virtual reality, and the method comprises the steps: collecting three-dimensional point cloud data, carrying out the convolution operation through a convolutional neural network algorithm, extracting a three-dimensional space feature matrix, and constructing a graphic recognition model. Inputting original data of a to-be-processed graph into the graph recognition model, outputting a graph parameter set, matching a virtual scene model in a preset scene library, and adjusting parameters of the model; establishing a physical coordinate system according to the adjusted model, converting a to-be-processed graph into grid units through a grid division algorithm, and calculating displacement data of the grid units; and reading displacement data, generating a graphic processing instruction sequence and sending the graphic processing instruction sequence to a graphic processor, and outputting an imaging result by superposing texture information and a ray tracing effect on the grid units. According to the method, a complete processing chain from data acquisition to rendering output is established, and the automation level and accuracy of graphic processing are improved.

The invention relates to the field of computer graphics, reverse engineering and additive manufacturing, discloses a multi-modal fusion and physical constraint three-dimensional point cloud geometric modeling system and method, and aims at overcoming the defects that traditional three-dimensional reconstruction entity attributes are missing, key features are prone to being lost, semantic understanding does not exist, and process automation is low. The system comprises a point cloud preprocessing module, a semantic segmentation and recognition module, a mixed geometric reconstruction module, a multi-component assembly and physical constraint solution module and an entity model output module. The method comprises the following steps: denoising and sampling an original point cloud, realizing semantic and instance double segmentation through a Point Net + + network, endowing a geometry-function composite label, completing hybrid reconstruction through improved Poisson reconstruction, parameterized B-Rep fitting and standard component library matching, and outputting a multi-format engineering available model after physical constraint verification and automatic correction of a violation model. According to the method, automatic, high-fidelity and high-practicability conversion from the point cloud to the engineering-level entity model is realized, and the method is adaptive to multiple industrial application scenes.

The invention discloses an SPH fluid rendering method and system based on a Gaussian function, and belongs to the technical field of computer graphics, and the method comprises the steps: receiving SPH simulated fluid particle data, and constructing a particle connection graph based on a particle neighborhood relation; based on the particle connection graph, extracting local geometric features of each particle through a neural network and aggregating the local geometric features into global geometric features; analyzing the global geometric features into Gaussian function parameters to generate a plurality of Gaussian functions for each particle, and superposing all Gaussian functions in space to form an implicit fluid surface; performing screen space rendering on the implicit fluid surface to obtain a final fluid rendering image; and after foam particles generated by SPH simulation are also mapped into a Gaussian function, a fluid rendering image containing a foam visual effect is obtained through screen space rendering. The method can improve the smoothness and reality of the rendering result while guaranteeing the calculation efficiency, and achieves the consideration of the efficiency and quality of fluid rendering.

The invention relates to the technical field of computer graphics and artificial intelligence, and discloses a meta-universe digital human generation method and system based on deep learning, and the method comprises the steps: constructing a neural radiation field human body model through multi-view video data, and learning a user behavior sequence mode; collecting sparse sensor data of the head and the hands of the user and generating a preliminary whole-body skeleton posture; calculating contact force feedback when the digital human interacts with the virtual environment through a differentiable physical engine; utilizing a reinforcement learning algorithm to optimize postures to meet physical constraints; performing multi-objective optimization and quality evaluation on the generated attitude; according to the method, the digital human postures which have personalized features and can naturally interact with the environment can be generated by using the consumer-level equipment, and the meta-universe interaction experience is improved.