1041 results about "Real-time rendering" patented technology

The invention relates to the technical field of communication command, in particular to an intelligent emergency command system and method based on scene twinning. Comprising a data acquisition module, a scene twinborn management module, an emergency situation analysis module, an emergency response decision module and a resource scheduling command module. Multi-source data of an emergency scene are collected through a sensor network, a scene twinning model is constructed by using augmented reality and digital twinning technologies, and real-time rendering of a virtual scene is realized. According to the method, emergency situation analysis, risk detection and trend prediction are performed in combination with a random matrix theory and a graph neural network algorithm, an emergency response strategy is generated by using a decision tree based on an analysis report, resources are scheduled and allocated through a multi-objective optimization algorithm, and an optimal emergency command scheme is ensured. According to the invention, the efficiency and scientificity of emergency command are improved, and the rapidity and accuracy of emergency response are enhanced.

Disclosed in the present invention is a dynamic human modeling method based on three-dimensional Gaussians. In the present method, a multi-view human video is given firstly, and in order to correct and transform the dynamic human appearance, a latent code and a set of blend weights are added to each Gaussian; and the latent code and a target pose are inputted to an MLP to correct the Gaussian in a canonical space, so as to capture an appearance change under the target pose. The corrected Gaussian is transformed into the target pose by means of the blend weights thereof in a linear blend skinning (LBS) manner. Finally, a realistic human image at a new viewing angle or a new pose can be rendered in real time by means of Gaussian splatting. Compared with an existing method based on an implicit neural radiance field (NeRF), the three-dimensional Gaussian representation of the present invention can better capture high-frequency details while achieving optimal rendering performance.

A method of constructing a digital twin physical field of a computer numerical control machine tool with double order reduction is provided. A physical field finite element simulation sample data set in a working state sampling space has an order preliminarily reduced to sparse grid node data via an improved K Nearest Neighbor (KNN) algorithm, so that the number of point sets is kept within the real-time rendering capability range of graphic rendering software. A digital twin physical field proxy model is trained using a radial basis function (RBF) to realize reconstruction of the physical field. The proxy model is iteratively optimized using a genetic algorithm to solve the problem of deviation between a proxy physical field model and a real physical model. Finally, the proxy model is deployed to a digital twin system, and the physical field distribution is solved in real time.

The invention discloses a multi-source three-dimensional model lightweight processing and lossless compression supporting method, system and device and a storage medium, and relates to the three-dimensional technical field, and the method comprises the following steps: carrying out format compatibility check on an imported three-dimensional model, and analyzing the geometric structure, texture characteristics and material attributes of the model; the geometric structure of the three-dimensional model is simplified by adopting a quadratic error measurement algorithm and edge folding, the simplified model is optimized through edge segmentation operation, and a lightweight model is obtained; and compressing geometric data of the lightweight model by using a data compression algorithm, and compressing texture data through predictive coding to complete lossless compression of the three-dimensional model. According to the method, the loading speed and the real-time rendering efficiency are remarkably improved by reducing the size of the three-dimensional model file and optimizing the texture compression algorithm. The lossless compression technique ensures that a high-quality visual effect is kept in the lightweight process, the lightweight model reduces the storage space requirement, and the energy consumption of data transmission and processing is reduced.

The invention relates to the technical field of data processing, and discloses a training method and system for generating a 5D digital human based on AIGC, and a medium. The method comprises the following steps: preprocessing real person feature data to obtain multi-dimensional preprocessed data; performing multi-level processing on the multi-dimensional preprocessing data to obtain 5D digital human generation data; performing cross-modal coordination processing on the 5D digital human generation data through a progressive multi-task learning algorithm to obtain to-be-processed digital human data; performing dynamic knowledge integration processing on the to-be-processed digital human data to obtain adaptive digital human data; carrying out personalized fine tuning and style migration processing on the adaptive digital human data to obtain target digital human data; and performing real-time rendering and intelligent interaction processing on the target digital human data to obtain target 5D digital human data, and generating a digital human training scheme according to the target 5D digital human data. According to the method and the device, the training efficiency and accuracy of generating the 5D digital human based on the AIGC are improved.

A high-quality digital human avatar is quickly reconstructed from a monocular video and is rendered in real time, the method is widely applied to various fields, and an existing method based on a neural radiation field is high in training and rendering cost. According to the dynamic human body three-dimensional modeling method based on the three-dimensional Gaussian sputtering technology, three-dimensional Gaussian can be expanded to a dynamic human scene, and rapid training and real-time rendering can be achieved. Specifically, the three-dimensional Gaussian of the human body is initialized in a standard space, and the three-dimensional Gaussian is converted from the standard space to a posture space by utilizing linear hybrid skin (LBS) according to posture parameters, so that fine details of the human body are learned at very low calculation cost; a posture parameter refining module and an LBS weight optimization module are designed to improve the problem of inaccurate detection of posture parameters in a real data set, and multi-resolution hash coding is introduced to accelerate the model training speed.

The invention provides a game cartoon character display method based on a virtual reality technology. The method comprises the following steps: acquiring three-dimensional model data of a game cartoon character in a target display area through a user interaction terminal; the virtual reality content management platform determines a dynamic rendering precision level according to the model data, and generates a real-time rendering strategy including a model patch reduction coefficient, a texture compression rate and a skeleton animation updating frequency in combination with terminal performance parameters; the strategy is sent to a virtual reality supervision platform and a user interaction terminal, and a virtual reality rendering engine platform is instructed to execute real-time rendering; and the virtual reality supervision platform monitors the frame rate fluctuation data of the head-mounted display device, calculates the scene rendering stability, sends a rendering optimization instruction if the scene rendering stability is lower than a threshold value, and adjusts the model data acquisition frequency and the video memory cleaning period to optimize the performance. The rendering efficiency and the system stability can be improved, and the hardware load and the frame rate fluctuation are reduced.

The invention provides a virtual human design and application platform, method and device based on artificial intelligence, and relates to the technical field of virtual digital humans. The method comprises the steps of performing local anonymization on multi-modal input data on user equipment, encoding generated anonymized multi-modal features to obtain a multi-modal feature vector, and inputting the multi-modal feature vector into an emotion calculation model to obtain a user emotion intensity quantized value; inputting the multi-modal feature vector into a context sensing model, and generating a user intention vector after context correction in combination with a knowledge graph; generating an updated personality parameter matrix according to the user emotion intensity quantized value and the user intention vector; and outputting voice waveform data, facial muscle motion parameters and skeleton joint coordinate data based on the personality parameter matrix, and driving the virtual digital human three-dimensional model to perform real-time rendering. According to the scheme, the naturalness, emotional resonance and long-term user retention rate of virtual digital human interaction can be improved, and user privacy data security is protected.

The invention provides a WEB browsing real-time rendering method and system based on distributed rendering, and relates to the technical field of distributed rendering, and the method comprises the following steps: receiving a rendering request from a user terminal browser, obtaining page data of a target webpage according to the rendering request, preprocessing the page data, analyzing the preprocessed page data, and displaying the analyzed page data. Obtaining a DOM tree structure and a CSSOM tree structure of the page, and constructing a rendering tree adaptive to the user terminal according to the DOM tree structure, the CSSOM tree structure and the equipment parameters; dividing the rendering tree into a plurality of sub-rendering trees, transmitting sub-rendering results generated by the plurality of distributed rendering nodes to a synthesis service node, performing image compression, transmitting the compressed rendering result data to a user terminal browser, receiving and decompressing the rendering result data in the user terminal browser to obtain a rendering result bitmap, and displaying the rendering result bitmap in the synthesis service node. And displaying the information to the user for real-time interactive browsing of the user.

The invention discloses a three-dimensional scene reconstruction method and device based on large model geometric prior, and a medium, and aims to solve the problems that a conventional 3DGS is liable to have artifacts and detail loss in geometric discontinuity, data redundancy and illumination variation scenes, and predicts a dense depth map and a normal map from a monocular image by using a pre-trained large model. The position and form of the Gaussian kernel are constrained as additional geometric priori; a primitive adjustment strategy based on kernel density estimation is introduced in the training stage, small Gaussian primitives with similar structures and adjacent spaces are combined into a large Gaussian primitive, the rendering quality is kept, redundancy is reduced, and the volume of the model is reduced; an exposure coefficient is adaptively estimated for each input image, an exposure compensation image loss function is constructed, and floating artifacts caused by illumination differences at shooting moments are eliminated. Experiments show that compared with the prior art, the method improves the three-dimensional reconstruction precision and real-time rendering quality of complex illumination and less-texture areas in a public data set and an unmanned aerial vehicle aerial photography scene.

The invention provides a game scene optimization method and system based on dynamic rendering, and belongs to the technical field of game development.The game scene optimization method comprises the steps that firstly, a real-time rendering state data set of a current game scene, rendering resource occupation information containing scene elements and picture frame rendering time consumption parameters are obtained; then, a pre-trained rendering demand analysis model is called for analysis, a rendering demand feature set of the scene elements is obtained, the rendering demand feature set comprises element visual importance features and rendering resource sensitivity features, and a scene element dynamic optimization strategy is generated based on the rendering demand feature set; the scene element dynamic optimization strategy comprises an element detail level adjustment rule and a rendering resource allocation priority parameter, executing a rendering parameter adjustment operation on a target scene element according to the scene element dynamic optimization strategy, generating adjusted scene rendering configuration data, and finally inputting the scene rendering configuration data into a game rendering pipeline for real-time rendering processing. And outputting the optimized game scene picture frame sequence. Therefore, the game picture quality and the operation performance can be improved.

The invention discloses a high-resolution three-dimensional reconstruction method of a fusion diffusion model, which belongs to the technical field of image data processing, and comprises the following steps: constructing an original data set D; constructing an enhanced training set; constructing a three-dimensional reconstruction network which comprises a text encoder, a renderer, a VAE encoder, a conditional diffusion model, a VAE decoder and an MVS module; training and fine-tuning the conditional diffusion model in three stages to obtain a three-dimensional reconstruction model, acquiring an image sequence and a text instruction of a scene to be reconstructed, and performing reconstruction by using the three-dimensional reconstruction model. According to the method, highly consistent geometric and color reduction can be kept under the multi-view condition, and splicing artifacts are remarkably reduced. Through semantic guidance optimization, texture details and structural consistency of the reconstruction model are greatly improved. Conditional diffusion sampling enables the model to accurately restore local details in a complex scene, and the stability of real-time rendering is improved.

The invention discloses a design scheme confirmation method and system based on a 3D model, and relates to the technical field of furniture design generation, and the method comprises the steps: collecting preference data of a user for furniture design to generate a two-dimensional hand drawing, recognizing the two-dimensional hand drawing through an image recognition technology, and generating a three-dimensional furniture model; rendering the three-dimensional furniture model, performing real-time modification according to user feedback to generate an effect picture, and generating a color matching scheme of furniture according to user preference data; and combining the three-dimensional furniture model with the user room live-action picture through an AR virtual reality technology to perform effect display. By combining the three-dimensional furniture model and the virtual illumination model, efficient real-time rendering in a complex environment is achieved, the illumination condition and the material reflection effect in a user room can be dynamically simulated, the rendering effect of the three-dimensional furniture model is more vivid, the confirmation efficiency and the visual precision of a design scheme can be remarkably improved, and the user experience is improved. The personalized requirements of the user are met, and the real-time response speed to the requirements of the user is increased.

The invention discloses a detail level rendering method based on Gaussian splash, which can dynamically change the detail level of a model according to the actual needs of a user in an interaction process, realize high-fidelity and low-resource-loss rendering based on Gaussian splash, and reduce the loss required by rendering. The technical scheme of the invention comprises a preparation stage and a real-time rendering stage. In the preparation stage, a multi-view image is acquired, a corresponding three-dimensional model or four-dimensional model is constructed according to the multi-view image, and Gaussian splash models with different detail levels are generated. And in the real-time rendering stage, the Gaussian splash models of different detail levels are selected for real-time rendering according to the real-time data of the user by using the Gaussian splash models of different detail levels.

The invention discloses a dynamic real-time rendering method for a large-scale assembly scene based on three-dimensional Gaussian splashing, a three-dimensional Gaussian splashing model for real-time rendering of a dynamic scene is constructed through combination of a plurality of sensors, and the method can render any view angle of the assembly scene. The three-dimensional Gaussian splash model comprises sensor data alignment, initialization of the three-dimensional Gaussian splash model and prediction and updating of the three-dimensional Gaussian splash model, and a scene space point alignment and predictor is designed by combining the three-dimensional Gaussian splash model, so that the problems of redundancy of multi-sensor data information and difficulty in dynamic real-time rendering are solved; according to the method, the assembly safety and reliability are improved, the problems in the assembly process can be found and processed in time, rapid and high-quality development of industrial intelligent manufacturing is promoted, and therefore transformation of industrial manufacturing intelligence is promoted.

The invention discloses a scene building method and system based on three-dimensional reconstruction. The method comprises the following steps: collecting a multi-view two-dimensional image of an environment; preprocessing the obtained two-dimensional image, and automatically identifying and labeling different objects and areas in the image by applying a semantic segmentation technology to understand semantic information in the image; according to the marked two-dimensional image data, reconstructing a three-dimensional model from the two-dimensional image by using a multi-view geometric three-dimensional reconstruction algorithm based on bundle adjustment, and storing the reconstructed three-dimensional model as a sub-scene model; according to the method, the three-dimensional model is accurately reconstructed by collecting the multi-view two-dimensional image and applying the semantic segmentation and bundle adjustment technology, the technical bottleneck of real-time rendering of a complex scene is solved, and the scene building efficiency and precision are improved; and in combination with the generative adversarial network and the variational auto-encoder optimization model, authenticity, detail expressive force and diversity are enhanced, and rich and real scene experience is provided for the field of virtual reality.

The invention provides a dynamic scene reconstruction method and system based on spatial decomposition and Gaussian splashing, and belongs to the technical field of computer vision and graphics. The method comprises the following steps: decomposing a dynamic scene into a static standard space and a dynamic playground; reconstructing the standard space by using three-dimensional Gaussian splashing to obtain Gaussian primitives of the standard space; decomposing the characteristics of the motion field into standard space characteristics and a plurality of motion subspace characteristics based on tensor decomposition; fusing the standard space feature and the multiple motion subspace features through a motion information decoder, and decoding motion information of Gaussian primitives in the standard space; and applying the motion information to Gaussian primitives of a standard space, generating any moment representation of a dynamic scene, and rendering a target view angle image through a snowball throwing method to realize reconstruction of the dynamic scene. Through the method, the reconstruction precision and the rendering quality are effectively improved, and the real-time rendering capability is kept.

The invention discloses a method for rendering a large-scale point cloud at a web end, and belongs to the technical field of large-scale point cloud rendering, and the method comprises the steps: S1, carrying out the data preprocessing of point cloud data, and carrying out the octree division of the whole point cloud; s2, on the basis of octree division, performing Lod hierarchical data division on leaf nodes of the whole octree; s3, data which needs to be rendered currently are screened out from the octree through view cone cutting; s4, high-level data and low-level data are separated from the screened data needing to be rendered, the low-level data are used for real-time rendering, and the high-level data are used for non-real-time rendering; and S5, combining the real-time rendering result and the non-real-time rendering result, and outputting all rendering results. According to the method, a similar double-buffer structure is used, and large-scale point clouds can be rendered in a real-time and non-real-time mixed mode at a web end.

The invention provides a digital human generation method based on a multi-modal large model. The method comprises the following steps: constructing a digital human basic model; generating a structured training set; generating a question and answer model supporting multi-channel interaction; semantic answers of the user questions are output, text emotional tendencies of the semantic answers are extracted, and emotional intensity parameters are output; generating facial muscle movement track data, and performing real-time rendering on the digital human basic model according to the facial muscle movement track data to output a digital human three-dimensional image with emotion expression. According to the embodiment of the invention, cross-modal alignment is carried out on text, image and audio data, and a multi-modal large model containing visual, voice and knowledge models is optimized by using a joint training method, so that more natural and smoother multi-channel interaction experience is realized; in addition, by introducing an emotion recognition model and a face interaction model, the emotion tendency contained in the semantic answer can be captured and reflected more accurately, so that a digital human three-dimensional image with real emotion expression is output.

The invention relates to the field of graphic rendering, in particular to a high-efficiency graphic rendering system based on artificial intelligence core computing power, which comprises an intelligent task decoupling module for performing task decoupling on a graphic rendering task; the AI enhanced texture generation optimization module is used for learning a dynamic scene change rule to generate a texture map in real time, reducing calculation overhead in a repeated texture region by using a multiplexing algorithm, and performing super-resolution compensation on low-resolution textures by using a deep convolutional neural network; the AI driving illumination and shadow calculation module is used for predicting an illumination transmission path in a scene by using AI and reconstructing a dynamic shadow effect so as to realize high global illumination and shadow simulation; the task result synchronization module is used for synchronizing rendering data output by each subtask; according to the method, the high-quality texture map is dynamically generated, super-resolution compensation is carried out on the low-resolution texture, repeated calculation is avoided through a texture multiplexing algorithm, the calculation overhead is remarkably reduced, and the real-time rendering efficiency of a large-scale complex scene is improved.

The invention discloses a cross-platform GPU support-free real-time digital human rendering system and method, and relates to the technical field of digital human rendering, and the system comprises an offline modeling module which constructs a mouth base space model through principal component analysis based on video data of a target person; the real-time driving module is used for mapping real-time voice features into dynamic coefficients of the mouth base space model; the image synthesis module is used for reconstructing a mouth image according to the dynamic coefficient and fusing the mouth image with a pre-stored reference face image; the cross-platform adaptation layer is configured to realize real-time rendering without GPU dependence through compiling, memory mapping and multi-thread scheduling technologies; and the dynamic resource manager is used for optimizing resource occupation and ensuring stable operation in a multi-platform environment. According to the technical scheme, dependence on high-computing-power hardware and an ideal acoustic environment can be broken through, and a complete theoretical framework and an engineering implementation path are provided for large-scale popularization of a digital human technology in a lightweight terminal.

The invention belongs to the technical field of three-dimensional scene interaction, and particularly relates to a three-dimensional immersive content interaction method and system based on 3D Gaussian spattering. The method comprises the following steps: collecting a multi-view image and carrying out motion structure recovery to obtain a camera pose matrix; according to the camera pose matrix, performing three-dimensional reconstruction processing on the image data through a 3D Gaussian spattering algorithm to obtain a reconstruction model; performing behavior analysis on the historical viewport track of the user to obtain a visual angle prediction result; and according to the visual angle prediction result and the reconstruction result of the reconstruction model, video frame generation is carried out at a client through a real-time rendering engine, and a transmission result is obtained. According to the invention, the application of the 3DGS in the three-dimensional expression technology is realized, the rendering quality is improved, and the degree of participation and immersion of the user are improved.

The invention belongs to application of a virtual reality technology in the field of automobile exhibition, and particularly relates to an automobile exhibition interactive experience method based on virtual reality. Firstly, a virtual exhibition hall and a vehicle type three-dimensional model are constructed and optimized, so that a user checks the appearance and the internal structure of a vehicle in an immersive manner; recording browsing behaviors, and analyzing preferences through a recommendation algorithm to generate personalized vehicle model recommendation; after a user selects a vehicle, configuration selection and real-time rendering display can be carried out, test driving can be selected, and the system loads different driving scenes to simulate control experience; and finally, capturing a user attention area based on a line-of-sight tracking technology of a clustering algorithm to generate a visual heat map so as to identify preference key points, and providing data support for marketing and exhibition planning. According to the method, the reality sense of virtual display is improved through multiple technologies, interactivity is enhanced, test driving experience is enriched, user preferences are accurately captured, and the defects of an existing automobile exhibition display mode are effectively overcome.

The invention discloses rendering equipment based on three-dimensional Gaussian sputtering. The hardware architecture includes a memory interface, a pre-processing engine, a cardinality ordering engine, a rasterization engine, and the like. The preprocessing engine converts the three-dimensional Gaussian point into a two-dimensional representation and generates a key value pair containing a tile identifier and depth information; the cardinal number sorting engine performs staged efficient sorting through a parallel processing mechanism and a BRAM alternate working mode; and the rasterization engine performs rendering processing by adopting a full-pipeline design and an early-stage stopping mechanism. According to the method, parallel execution of depth sequencing and rasterization is achieved, memory access and calculation redundancy are reduced through optimized data flow design, an efficient and low-power-consumption hardware acceleration solution is provided for three-dimensional Gaussian sputtering rendering, and the method is particularly suitable for application scenes such as virtual reality, games and scientific visualization requiring real-time rendering.

The invention belongs to the technical field of augmented reality (XR) and computer vision crossing, relates to fusion application in immersive digital performance, and provides a virtual actor reconstruction and immersive presentation system based on 4D Gaussian splash modeling and XR space positioning. The system comprises a set of spherical multi-camera-position high-synchronization camera shooting matrix used for capturing dynamic images of actors; performing dynamic modeling on the multi-angle image through a 4D Gaussian splashing technology, and outputting a virtual actor point cloud model which can be used by XR equipment; vPS visual positioning and an SLAM tracking module are combined, and precise mapping positioning of a performance space is achieved in AR / MR equipment. The method supports the immersive watching of the actor image at the audience end at a 360-degree free visual angle, and realizes the natural presentation of the virtual actor without dead angles and wearing in cooperation with shielding judgment and a real-time rendering engine. The system is widely applicable to on-site entertainment scenes such as immersive theaters, text travel performances, brand activities, concerts and television programs.

The invention provides a virtual tourism experience method based on a large model, and aims to provide personalized and immersive tourism experience by integrating advanced artificial intelligence and virtual / augmented reality technologies. The method comprises the following steps: firstly, collecting interest and preference information of a user, such as historical travel records, scenic spot preferences, travel time preferences, social media activities and the like, so as to capture detailed demands of the user; thirdly, generating personalized tourist routes and suggestions according to the information by utilizing a large-scale artificial intelligence model and combining deep learning and natural language processing technologies; in addition, the system renders scenic spots in a tour route in real time through a virtual reality or augmented reality technology, and the reality sense of vision and hearing is improved by applying advanced graphic and physical simulation technologies. And finally, an interactive interface is designed, so that the user can interact with the virtual tour guide through voice, text or gestures, more scenic spot information is obtained, a tour route is adjusted in real time, and the individuation and participation sense of user experience are greatly enhanced.

The invention discloses a dynamic three-dimensional scene reconstruction and real-time rendering method based on free space-time Gaussian sputtering, and the method comprises the steps: firstly obtaining a multi-view-angle dynamic scene video, extracting cross-view-angle feature points for three-dimensional reconstruction, initializing free space-time Gaussian primitives, and carrying out the real-time rendering of the free space-time Gaussian primitives; an optimizable explicit motion function and a time opacity function are set for each Gaussian primitive and are used for representing the geometry and appearance of the dynamic three-dimensional scene; then sputtering and rendering the Gaussian primitive at the current moment based on the observation visual angle, and outputting a high-fidelity dynamic scene image of the corresponding visual angle; finally, Gaussian primitive parameters are optimized in a combined mode through a rendering loss function and a four-dimensional regularization strategy, and meanwhile low-influence Gaussian primitives are relocated periodically. According to the method, the movable Gaussian primitive is introduced at any position in space and time, the explicit motion function and the time opacity function are combined, a complex dynamic scene is effectively modeled, and efficient reconstruction and real-time rendering of the dynamic three-dimensional scene are achieved.

The invention relates to a rendering method for local loading cache synchronous optimization rendering of a large-scale basin twinborn model, and belongs to the field of digital twinborn model rendering. The method comprises the following steps of: reconstructing a top index for an original model data set, and combining the original model data set into an integral data set through vertex compression and quality compression so as to realize the purpose of accelerating rendering; according to the method, detail models of different levels are created for different areas, so that the real-time rendering precision of the area far away from an observer is reduced, the calculated amount is reduced, optimization of an overall rendering pipeline is carried out in combination with a fragment shader and a vertex shader, thread consumption of a GPU is reduced, and real-time loading of a large-scale drainage basin twin model and display of simulation scene change are achieved. According to the method, the large-scale digital twin model can be lightened, the overall data size of the model is reduced while the visualization effect is guaranteed, and meanwhile, rendering visualization of the large-scale twin model can be dynamically loaded in real time.

The application relates to a dynamic visualization method and device for earthquake disasters. The method comprises: acquiring initial point cloud data of buildings and scenes in a target area; acquiring shake displacement data varying with time of the buildings by simulating an earthquake excitation on the buildings and the scenes; performing voxel downsampling and point cloud segmentation and extraction on the initial point cloud data to determine a ground point cloud and a point cloud of each building; updating in real time coordinates of each point of the point cloud of each building according to the shake displacement data; and rendering in real time the ground point cloud and the coordinates of each point of the point cloud of each building to generate a dynamic visual image for displaying. Accordingly, seismic scenes of regional buildings are presented more truly and visually, thus improving the visualization effect.

The invention relates to the technical field of three-dimensional animation production, and discloses a three-dimensional animation production method based on AI deduction real-time rendering output, which comprises the following steps: S1, motion capture: in a motion capture stage, in combination with an AI technology, carrying out real-time analysis and optimization on original motion capture data by using a deep learning algorithm, and through a human body posture estimation technology, carrying out real-time analysis and optimization on the original motion capture data; the motion of the human body can be more accurately captured, and facial expressions and emotion changes are captured in real time in combination with an emotion recognition algorithm, so that more real and vivid animation role performance is provided; by optimizing and simplifying the three-dimensional animation production method, the production difficulty is remarkably reduced, the time is shortened, the cost is reduced, and children 'animations generally require high-frequency animation updating and high-quality visual effect to attract and maintain the interest of small audiences. The AI technology is integrated for action capture and emotion recognition, so that expressions and actions of roles can be captured and optimized in real time, and the roles can be expressed more naturally and vividly.