16 results about "Human Modelling" patented technology

The invention discloses a digital human rendering method based on Gaussian splashing and multi-scale characteristic field distillation, and belongs to the field of three-dimensional human body digital reconstruction. According to the method, feature extraction is carried out through a Vision Transformer encoder, based on an SMPL model, through a cross-modal parameter estimation module and dynamic human body modeling of three-dimensional Gaussian splashing and semantic feature rendering, three-dimensional Gaussian is projected to a two-dimensional image plane to calculate a covariance matrix and color mixing, and after a rendered color image and an initial feature field image are output, a three-dimensional image is obtained. A student feature map is obtained through a convolution acceleration module, a teacher feature map is obtained after feature extraction is carried out through a two-dimensional basic model, the constructed model is trained, and optimization training is completed through comprehensive total loss function calculation; according to the method, the problems of fuzzy semantics, detail missing, low rendering efficiency and inaccurate human body-scene separation in the existing method are effectively solved, so that more efficient, fine and robust monocular or multi-view human body three-dimensional reconstruction is realized.

The invention relates to the technical field of computer vision, and particularly provides a virtual digital human generation method and electronic equipment, and the method can comprise the steps: obtaining three-dimensional human body modeling data corresponding to each frame of image in a monocular video sequence of a target object; the three-dimensional human body modeling data comprises body posture data and identity offset data; processing the three-dimensional human body modeling data by using a pre-constructed to-be-trained virtual model to generate a virtual object image of each frame of image; the to-be-trained virtual model comprises a to-be-optimized three-plane feature body and a plurality of multi-layer neural network MLP models; optimizing the to-be-trained virtual model through the loss value between the virtual object image and each corresponding frame image, and obtaining a three-dimensional digital human model corresponding to the target object; the three-dimensional digital human model is used to generate a virtual digital human based on any drive data. According to the embodiment of the invention, the generation and rendering quality of the virtual digital human can be improved.

The application discloses a 3D human body modeling method and device based on contrast learning and a medium. The method comprises the following steps: acquiring 2D image information corresponding to a user and body state information corresponding to the user; extracting original features corresponding to the 2D image information through a shared encoder; performing mask prediction and determining clothing coverage rate according to the original features; performing feature fusion according to the original features, shape prior features corresponding to the body state information and the clothing coverage rate to obtain fused features; performing adaptive feature purification on the fused features according to the clothing coverage rate to obtain pure features; and performing human body parameter prediction according to the pure features.

The invention discloses a convenient three-dimensional human body reconstruction method based on a monocular video, and the method comprises the steps: constructing a parameterized human body model low-dimensional shape space, and obtaining the approximate representation of a linear model. And extracting a multi-view key frame from the input human body rotation video. And extracting a feature vector of each view key frame by using an improved ResNet-50 network fused with an attention mechanism. And defining the multi-view feature vectors as graph nodes, and constructing a graph structure. And performing cross-view depth feature fusion through the graph convolutional network to obtain global features. And mapping the global features to a parameterized human body model low-dimensional shape space. And reconstructing a complete three-dimensional human body grid by using linear model approximate representation. The method has the advantages that multi-view feature extraction and a graph structure fusion mechanism are combined, the end-to-end process from consumption-level video input to high-precision three-dimensional human body reconstruction is achieved, certain generalization ability is achieved in a real scene, and a low-cost and high-precision three-dimensional human body modeling solution can be provided for virtual fitting, digital twinning and other applications.

The invention discloses an intelligent fitting recommendation system based on 3D human body modeling and clothing parameter adaptation, and relates to the technical field of intelligent fitting recommendation. The system generates a high-precision dynamic body surface grid through a multi-modal data acquisition module; a human body-costume double-layer decoupling digital model construction module is utilized to realize efficient and realistic costume dynamic deformation simulation; through a dynamic pressure simulation and evaluation module, in combination with an improved Hertz contact theory and a fabric mechanical model, carrying out accurate simulation and multi-dimensional quantitative evaluation on dynamic contact pressure of clothes and a human body; and finally, the scene intelligent recommendation module provides personalized clothing recommendation according to the scene selected by the user and the dynamic evaluation result. According to the method, the problem that traditional virtual fitting lacks dynamic comfort evaluation is solved, the user experience is remarkably improved, the return rate of merchants is reduced, and the method has important market application value.

The invention discloses a shopping mall virtual fitting room implementation method and system based on the AR technology, and belongs to the technical field of e-commerce and computer vision, and the method comprises the steps: carrying out the real-time high-precision three-dimensional human body modeling based on a common camera; performing intelligent cloth physical simulation fused with a machine learning algorithm; multi-terminal cross-platform consistency rendering is supported; and carrying out personalized clothing recommendation in combination with user body type data. The virtual try-on effect of the user can be truly presented, multi-device terminal access is supported, the shopping experience is remarkably improved, and the refunding and changing rate is reduced.

The invention relates to the technical field of costume design, and discloses an intelligent interactive costume design system and method based on a data model. The system comprises an interaction module, a human body modeling module, a demand intelligent analysis module, an intelligent design generation module, a fabric matching module, a virtual fitting module and a scheme optimization module. The interaction module obtains PLY format human body data of a user and costume design requirements and receives modification feedback of the user, and the human body modeling module generates a 3D human body model, human body size parameters and posture feature tags. The demand intelligent analysis module converts a costume design demand into a structured demand document containing a style tag and a version constraint parameter in combination with a posture feature tag, the intelligent design generation module outputs an optimal costume design scheme through a VAE generation model, and the fabric matching module matches candidate fabrics based on the optimal costume design scheme. The virtual fitting module realizes dynamic fitting simulation, and the scheme optimization module iteratively optimizes the optimal costume design scheme to meet user requirements. The personalized costume design efficiency of the user is improved.

The invention discloses a three-dimensional human body modeling method and system based on binocular vision and a laser radar, solves the problem of poor reconstruction effect of a three-dimensional human body in a dynamic motion scene, and belongs to the technical field of computer vision, intelligent perception and three-dimensional modeling. A binocular camera and a laser radar are used for synchronously collecting a human body, and a binocular image and a laser radar point cloud are obtained; processing and fusing the obtained binocular image and the laser radar point cloud to generate a human body initial three-dimensional model; tracking and updating the parameterized human body model in real time, and outputting a dynamic three-dimensional human body model; in the real-time tracking and updating process, a parameterized human body model is constructed according to the human body initial three-dimensional model, the human body volume, the image contour and the point cloud distance are used as constraint terms, the human body elastic deformation is used as a physical regularization term to construct an optimized energy function, and the deformation of the parameterized human body model in motion is optimized by using the optimized energy function. According to the method, the geometric fidelity and the detail reduction degree of the three-dimensional model are effectively improved.

The present application relates to the technical field of virtual fitting, and more particularly to an intelligent virtual fitting method based on three-dimensional human modeling. The method comprises the following steps: collecting multi-angle image data of a human body and constructing a three-dimensional model of the human body structure to accurately capture various dimensions and postures of the human body, thereby providing accurate basic data for virtual fitting; obtaining a garment and collecting garment physical parameters to analyze the appearance effect of the garment; using the three-dimensional model of the human body to simulate postures and combining the physical characteristics of the garment to analyze the interactive dynamic response of the human body and the garment, thereby obtaining dynamic effect data of the garment under different postures; and further optimizing the virtual fitting effect by calculating the matching degree of the garment and the human body; the present application optimizes intelligent virtual fitting to achieve more perfect virtual fitting matching degree.

A human posture estimation method, device, equipment and storage medium are disclosed, and belong to the technical field of image processing. The method comprises the following steps: acquiring a first data set; training a three-dimensional human modeling model by using the first data set; and performing human posture estimation based on a three-dimensional human model output by the trained three-dimensional human modeling model. The three-dimensional human modeling model comprises a ResNet feature extractor, an HMR regressor and an SMPL model connected in sequence. The input of the HMR regressor comprises image features of a first image and boundary box geometric features of a second human body. The output of the HMR regressor comprises joint point features. The boundary box geometric features are used to convert the joint point features to a first coordinate system, so as to calculate a projection loss of the joint point features in the first coordinate system. The first coordinate system is a coordinate system of an original camera. The method can accurately realize human posture estimation on an image with slight posture changes.

The application discloses a three-dimensional Gaussian dynamic human body modeling method based on compressed three planes, and comprises the following steps: obtaining a camera pose parameter sequence, human body SMPL shape information and a posture information sequence from a single-view video sequence, and obtaining an initialized three-dimensional Gaussian point cloud; obtaining high-dimensional features according to the three-dimensional Gaussian point cloud, and predicting information of the three-dimensional Gaussian in a standard space; distilling skin weights of each vertex of a standard SMPL model and smoothing; generating high-dimensional position coding of the three-dimensional Gaussian by splicing, combining SMPL posture information and three-plane high-dimensional features, predicting basic attribute transformation values, and correcting Gaussian attributes under different human body postures; calculating a transformation matrix by using the corrected skin weights and the SMPL posture information, and obtaining a transformed three-dimensional Gaussian point cloud by transformation; rendering the transformed three-dimensional Gaussian point cloud, combining an image loss function and attribute constraints of neighbor Gaussians, and jointly optimizing to obtain a drivable human body three-dimensional Gaussian model. The application also discloses a system for realizing the above method, and has wide application scenarios.

This invention provides an integrated clothing customization system and method based on AI technology, belonging to the field of smart retail and e-commerce technology. The system adopts a dual-space architecture: the first space enables contactless AI body measurement, 3D human body modeling, and digital interaction; the second space automatically handles the storage and retrieval of physical sample garments. It integrates modules for size extraction, virtual try-on rendering, physical try-on feedback, and order generation. AI algorithms achieve accurate body measurement, simultaneous virtual and physical try-on, and parametric pattern generation. The method includes human body image acquisition, 3D modeling and size extraction, virtual try-on, physical try-on feedback, order generation, and production integration. This invention achieves integrated measurement, try-on, customization, and production processes, improving measurement accuracy and the try-on experience while shortening the production cycle.

The invention relates to the technical field of 3D human body modeling, and discloses an AI model-based 3D digital human building method, which comprises the following steps of: acquiring image data, action data and voice data of a target person, and respectively preprocessing the acquired data; constructing and training a 3D digital human modeling model, an action generation model and a voice driving model; and establishing a 3D digital human model according to the preprocessed data and the trained models. According to the method, multi-aspect data of the target person is comprehensively collected and preprocessed, different models are constructed for targeted training, and the 3D digital person model fitting the features of the target person can be accurately and comprehensively generated.

The application relates to the technical field of three-dimensional human body modeling, and particularly discloses a modeling and rendering method and device based on layered three-dimensional Gaussian representation and a medium, method steps comprising: acquiring an image sequence of a target, initializing three-dimensional Gaussian points in a canonical space based on SMPL human body template vertices; taking the initialized three-dimensional Gaussian points as the first layer on the innermost side, constructing a layered Gaussian structure from coarse to fine, and establishing conditional dependence between layers through a parent-child node connection mode; setting layered motion conditions, and respectively predicting rigid body and non-rigid body deformation transformation parameters of each layer of Gaussian points by using a rigid body transformation network and a non-rigid body transformation network; uniformly mapping the layered Gaussian points to an observation space according to the predicted transformation parameters, and performing image synthesis by using a three-dimensional Gaussian rendering technology. Compared with the prior art, the application can naturally adapt to and express deformation characteristics of different levels in human body motion, and realizes high-quality reconstruction and real-time rendering of a complex dynamic scene.

This invention discloses an intelligent video surveillance management method and system based on big data, relating to the field of video surveillance technology. The method includes using a VIBE model to generate SMPL-X model parameters to obtain the vertex coordinates of the head in a 3D human body mesh; using a DECA model to generate FLAME parameters to obtain the vertex coordinates of the face in a 3D face mesh; and fusing the vertex coordinates of the human and face meshes. By introducing a keypoint extraction method combining traditional image processing and deep learning, the accuracy of keypoint detection in occluded and low-light scenes is improved, enhancing the robustness of 3D human body modeling. Dynamic region masking and the YOLACT++ model are used to optimize the semantic feature extraction of moving regions. Through fixed-point coordinate fusion optimization, a high-precision 3D human body model is generated. Combined with encryption and multi-level LoP video output, it flexibly adapts to high-privacy and resource-constrained scenarios, improving the adaptability and practicality of the surveillance system.

An Augmented Reality (AR) and Artificial Intelligence (AI)-constructed, interactive three-dimensional (3D) human body modeling system that analyzes data descriptive of medical records and generates an interactive 3D human body model output via AR. The 3D human body model may have body part-specific interactive medical record data embedded therein, permitting a user to quickly, easily, and intuitively navigate the model to identify relevant medical status for the modelled individual.