190 results about "Polygon mesh" patented technology

Building on U.S. Provisional Patent Application No. 63 / 693,803, paragraphs for 2D-to-3D conversion and [0023]-[0024] for secure content verification, the Matrix 3D AI Polygon Mesh Hash Key System revolutionizes transforming 2D content into secure, interactive 3D AR / VR assets. Integrating AI and LiDAR, it enables real-time 3D mesh generation with precise geospatial anchoring. Users can reshape 3D content instantly via natural language commands, enhancing interactivity. Quantum-resistant encryption, using AES-256 and CRYSTALS-Kyber, ensures robust asset security. The system excels in gaming, surveillance, content verification, military operations, space exploration, deepfake detection, and pharmaceutical research, offering unmatched precision and scalability. Its multi-stage rendering pipeline, powered by distributed AI-driven bots, supports efficient real-time 3D mesh generation, achieving 95% accuracy and 1 cm resolution. This AR / VR technology advancement transforms industries with scalable, secure solutions for interactive 3D content creation and management, setting a new standard for precision and versatility.

The invention discloses a transparent object inverse rendering method and device based on three-dimensional Gaussian, and the method comprises the steps: firstly collecting a multi-view image in a three-dimensional scene with a transparent object, and initializing a three-dimensional Gaussian primitive; dividing the Gaussian primitive into a transparent object Gaussian primitive and an environment Gaussian primitive, baking based on an optimal projection strategy to obtain an environment Gaussian light field probe, calculating based on a bidirectional scattering distribution function carried by the transparent object Gaussian primitive to obtain a corresponding attribute map, and calculating reflected light and refracted light through an iterative probe query algorithm to obtain a target object. Transparent object reconstruction is realized through loss optimization of parameters of Gaussian primitives of the transparent object; and finally, based on the optimized transparent object Gaussian primitive, extracting a polygon mesh, recombining the polygon mesh with the environment Gaussian primitive to be recombined to obtain a recombined scene, and re-rendering the recombined scene. According to the method, performance advantages are kept, and meanwhile, the method also has advantages in inverse rendering and new view angle synthesis quality.

A phononic crystal generative design method based on a diffusion model and an unstructured polygonal mesh belongs to the field of phononic crystal design, and comprises the following steps: firstly, generating phononic crystal unit cells, calculating a dispersion curve of the phononic crystal unit cells, and constructing a data set; secondly, constructing a diffusion model and designing a neural network to learn a reverse process of the diffusion model; inputting the training data into the neural network for training until a termination condition is met; and finally, extracting a dispersion curve from the test data set as a reverse design target, quickly generating a group of photonic crystals by using a neural network and a diffusion model sampling algorithm, calculating the dispersion curve corresponding to the generated photonic crystals, and calculating an error between the dispersion curve and the reverse design target. According to the method, the dependence of an existing generative design method on a rule design domain can be broken through, and the adaptability to irregular geometrical shapes and complex boundaries is improved; efficient learning of the complex mapping relation between the mechanical attributes and the geometric configuration is achieved; the method has the advantages of high design speed and high precision.

Some aspects of the disclosure provide a method of mesh decoding processing. The method includes receiving a bitstream including coded information of a polygon mesh, the polygon mesh including vertices that are connected into polygons, the coded information indicates connectivity information of the vertices. The method also includes determining a first connectivity of a first polygon-fan with a first vertex being a pivot vertex of the first polygon-fan according to the coded information; detecting that a second vertex in the first polygon-fan is a a visited vertex that has existing neighborhood information; checking whether the existing neighborhood information of the second vertex and new neighboring information of the second vertex in the first polygon-fan satisfy a condition; and updating the existing neighborhood information of the second vertex based on the new neighboring information when the condition is satisfied.

In a method, a bitstream of a mesh that includes a plurality of vertices is received. When the mesh is symmetric with respect to a symmetry plane, a first group of vertices of the mesh that is positioned at a first side of the symmetry plane is reconstructed. A second group of vertices of the mesh that is positioned at a second side of the symmetry plane is reconstructed by reflecting the reconstructed first group of vertices with respect to the symmetry plane.

The invention discloses a morphology reconstruction method for a deposition layer of a ceramic matrix composite air film cooling structure, and belongs to the field of engineering thermophysics. According to the method, on the basis of traditional fluid-solid coupling heat transfer simulation, a track of dust particles on the surface of the ceramic-based composite material air film cooling structure is simulated in combination with a dispersed phase model, deposition layer thickness distribution characteristics are obtained, and after coordinate positions and normal vectors on surface grids are obtained through a user-defined function, the deposition layer thickness distribution characteristics are obtained. And directionally moving the coordinates at different positions of the component along the normal direction by the distance of the deposition thickness to obtain coordinate values of the surface of the sediment, and sequentially reconstructing the obtained coordinate point clouds into a polygonal mesh surface and a geometric model to obtain the geometric morphology of the reconstructed deposition layer. Therefore, the comprehensive cooling efficiency and attenuation characteristics of the ceramic-based composite material air film cooling structure under the influence of the deposited particles can be accurately calculated by simulating and analyzing the influence of the deposited layer on parameters such as flow characteristics and temperature distribution on the wall surface.

A bitstream that includes coded information of a mesh is received. The mesh includes a plurality of vertices. Whether a current vertex of the plurality of vertices is a pivot vertex is determined based on a priority of the current vertex in the plurality of vertices. The current vertex includes a plurality of incident faces. The priority of the current vertex is determined based on a total number of already visited incident faces of the plurality of incident faces of the current vertex. One or more unvisited incident faces of the plurality of incident faces of the current vertex are reconstructed when the current vertex is determined as the pivot vertex.

Generating polygon meshes approximating surfaces with sub-cell features. In some implementations, a computer-implemented method includes obtaining a signed distance field (SDF) grid that includes a plurality of cells, the cells including cell values that indicate distances of the cells to a surface that distinguishes an inside and an outside of an object. A boundary mesh is determined having boundary vertices and boundary faces of particular cells in the SDF grid that are based at least on cells within a threshold distance of the surface. Offset cell positions are determined for centers of neighboring cells that neighbor the boundary mesh. The offset cell positions are based on corner gradients of the neighboring cells. An adjusted mesh is generated that approximates the surface, the adjusted mesh defined by mesh vertices that are based on the boundary vertices of the boundary mesh that are displaced based on the offset cell positions.

Aspects of the disclosure includes methods and apparatuses for coding connectivity information of a polygon mesh. A method for decoding connectivity information of the polygon mesh includes receiving coded information including a first syntax element of a split vertex incident to a current face that is being processed. The current face is one of a plurality of faces of the polygon mesh. The method includes determining a number of edges between a pivot vertex incident to the current face and another vertex incident to the current face. Each of the edges between the pivot vertex and the other vertex is incident to a respective processed face of the plurality of faces. When the number of edges between the pivot vertex and the other vertex is indicated by the first syntax element, the method includes determining that the other vertex is the split vertex.

A polygon mesh includes a plurality of vertices that are connected into faces, a first face in the faces includes at least a first vertex, a first previous vertex of the first vertex, a first next vertex of the first vertex in a first list of vertices that are incident to the first face. A method of mesh processing includes generating a first incident face set associated with the first vertex, the first incident face set includes a first set of faces that includes the first vertex as an incident vertex, the first incident face set including the first face. Further, the method includes configuring an array according to the first incident face set associated with the first vertex, and determining an iteration order for processing the first set of faces based on the array.

The present invention relates to a method for generating a three-dimensional digital model of a drill core, comprising the steps of positioning at least one drill core fragment in an acquisition scenario; capturing images of the drill core fragment; generating a mask for each captured image; reconstructing the position and angle of the image; generating at least one dense point cloud; generating at least one polygonal mesh; georeferencing the drill core fragment; performing texture processing; and unifying a plurality of drill core fragments into a single three-dimensional object.

The invention discloses a three-dimensional object reconstruction method based on interactive segmentation prior and implicit surface reconstruction, which interactively extracts a semantic segmentation mask and a target image of a target object based on a pre-trained SAM segmentation model, and introduces an SfM algorithm to calculate camera attitude information and sparse structure information of a multi-view scene graph. And inputting the target image, the segmentation mask and the camera attitude information into a grid encoder for training to obtain a rough polygon grid and a texture feature, reconstructing the rough polygon grid and a texture map of the target object, and outputting a three-dimensional reconstruction result of the target object after refining the rough polygon grid and the texture map. According to the method, the interactive segmentation method is constructed through the pre-training large model, so that invalid occupation of a background region is avoided. The implicit surface reconstruction algorithm based on the neural radiation field is used for constructing the fine grid and texture information of the object to be reconstructed, three-dimensional reconstruction of the target object can be rapidly and accurately completed in a complex environment, and the method has good value and popularization prospects.

A method includes generating a bitstream comprising an encoded three dimensional polygon mesh in accordance with dual degree connectivity comprising, when the polygon mesh comprises at least two different face degrees, a first sequence representing a vertex degree of each vertex in the polygon mesh, and a second sequence representing a face degree of each face in the polygon mesh, where each vertex degree in the first sequence and each face degree in the second sequence is followed by a degree offset, and at least one degree corresponding to a vertex degree in the first sequence or a face degree in the second sequence is encoded in accordance with a context adaptive binary arithmetic coding (CABAC) model that encodes the at least one degree using a difference between the at least one degree and a degree mode that indicates a number of different types of face degrees.

This disclosure provides methods, devices, and systems for planning and performing medical procedures. The present implementations more specifically relate to analyzing objects in 3D images. In some aspects, a segmentation system may receive image data representing a 3D image of an anatomy, select a seed location for a target in the 3D image, and infer a segmentation mask associated with the seed location from at least a portion of the received image data based on a neural network model trained to segment a class of objects associated with the target. The system further extracts a polygon mesh from the segmentation mask to produce a 3D model of the target. The system can determine a spatial relationship between an instrument and the target based on a position of the 3D model relative to the 3D image. The system can also estimate a geometry of the target based on the 3D model.

An example of an information processing system stores first volume data that is data indicating a terrain object and includes a plurality of pieces of voxel data in a first voxel space, and second volume data that is data indicating an enemy object and includes a plurality of pieces of voxel data in a second voxel space. If a destruction action on the terrain object is performed, voxel data of a voxel included in a first range in the first volume data is updated. If a destruction action on the enemy object is performed, voxel data of a voxel included in a second range in the second volume data is updated. Based on the first volume data and the second volume data, a polygon mesh is created.

Aspects of the disclosure includes methods and apparatuses for coding connectivity information of a polygon mesh. A method for decoding the connectivity information includes selecting an active vertex from a plurality of active vertices in a plurality of vertices as a pivot vertex that is to be processed based on a priority of the active vertex. The priority of the active vertex is based on one of (i) a number of processed faces to which the active vertex is incident, (ii) a sum of an angle of each processed incident corner of the active vertex, (iii) a number of unprocessed faces to which the active vertex is incident, and (iv) a number of unprocessed face fans to which the active vertex is incident. The connectivity information of the polygon mesh is reconstructed using the selected active vertex as the pivot vertex.

Apparatuses, systems, and techniques to generate a 3D mesh using one or more neural networks. In at least one embodiment, tokens of a sequence of tokens that represents a sequence of coordinates of vertices of polygons of a polygon mesh are generated autoregressively using one or more neural networks based on a 3D representation of an object and one or more attributes. In at least one embodiment, the sequence of tokens includes start-of-part and end-of-part tokens that separate vertex coordinates corresponding to a part of the 3D representation of the object. In at least one embodiment, the one or more neural networks are trained via a progressive training technique that employs iterative vocabulary expansion and fine-tuning.

The invention belongs to the technical field of three-dimensional modeling, and discloses an effective building model Low-poly generation method based on re-gridding, and the method comprises the steps: carrying out the preliminary feature extraction of an input original grid Mi, and generating unsigned distance field data and feature data; feature mapping guides selection of re-gridding, feature enhancement simplification and polygon triangularization. In the re-gridding process, the grid is simplified as much as possible, meanwhile, features are reserved, the grid quality is ensured, and finally a reconstructed grid Md is generated; and according to the number of the target patches, further applying a feature preserving algorithm to simplify the Md, and generating a final low polygon mesh Mo. According to the method, the quality problem existing in the original grid can be solved under the condition that the original grid is closer to, and the conditions that the grid with the minimum patch number in the LODs is generated, and the patch number can be controlled by a user (Low-poly) are met; the visual appearance of Mo is similar to that of Mi at any long distance; mo is a watertight, manifold, and intersection-free high quality grid.

A neural radiance field (NeRF) for rendering an image in response to a text description receiving a text description is generated by using first and second polygonal meshes to establish spatial constraints. Points of the NeRF are scored using the spatial constraints to modify the NeRF, which may then be used, typically after conversion to a mesh, in rendering a computer simulation character or object.

According to an aspect of the present disclosure, a method of generating a low polygon mesh of a three-dimensional (3D) model is provided. In the method, an edge folding operation is performed on each of a plurality of edges of an input mesh to generate a simplified mesh, where the number of faces of the input mesh is reduced by the edge folding operation. The position of at least one vertex of the simplified mesh is updated to generate an updated mesh where the distance between the simplified mesh and the input mesh decreases. A low polygon mesh is generated based on the aligned mesh, where the positions of vertices of the updated mesh are optimized by minimizing shape differences between the simplified mesh and the updated mesh.

The invention relates to the technical field of three-dimensional vision, in particular to a three-dimensional model construction method and system for image registration matching, which can utilize a depth camera to obtain image data and depth data of a target at a high visual angle and a low visual angle, realize space-time alignment and generate point cloud data; performing feature extraction on the grayscale image data, and generating descriptors according to gradient features of areas around the key points; carrying out feature matching on the key points to obtain a plurality of matching point pairs, and carrying out mixed superposition on the color images to generate a superposed image; and calculating the normal vector and curvature of each point cloud surface, converting discrete point cloud data into a continuous three-dimensional surface, and constructing a three-dimensional polygonal mesh model. According to the technical scheme, under the condition that hardware parameters of the depth camera are not changed, information of multiple visual angles is fused, the data fusion problem is effectively solved, the model reconstruction precision and details are improved, the development threshold is lowered, and efficient application of the three-dimensional visual technology in multiple fields is achieved.

A method of mesh processing includes receiving coded information of a mesh, the coded information including position connectivity of a plurality of 3D vertices of a mesh in a 3D space, and a correspondence of the plurality of 3D vertices to UV vertices in a UV space for the mesh; determining UV connectivity of the UV vertices; determining, based on the coded information, a first initial UV vertex in a first UV chart in the plurality of UV charts; determining first UV vertices in the first UV chart according to a traversing order of 3D vertices corresponding to the first UV vertices in the first UV chart, the traversing order starting from the first initial UV vertex and traversing each UV vertex in the first UV chart; decoding, from the coded information, prediction residuals of the first UV vertices; and determining first UV coordinates of the first UV vertices accordingly.

A computing device for displaying a 3D virtual object for an augmented reality (AR) scene is described. The computing device has a processor and a non-transitory computer-readable memory, wherein the processor is configured to carry out instructions from the memory that configure the computing device to: determine a polygon mesh representation of a target object within the AR scene, wherein the polygon mesh representation models a physical appearance of the target object; identify a target mesh surface of the polygon mesh representation onto which the 3D virtual object should be anchored; monitor the target object and update the polygon mesh representation according to changes in the physical appearance of the target object; and display the 3D virtual object within the AR scene according to a position of the target mesh surface on the updated polygon mesh representation, wherein the target mesh surface moves along with the target object.

The invention relates to the technical field of medical image calculation, in particular to a heart three-dimensional polygon mesh construction method and system, and the method comprises the steps: firstly obtaining and registering an image, generating an inductor density and a signed distance field based on boundary features, and fitting anatomic anchor points; determining a patient specific three-dimensional domain through mass conservation transmission and homeomorphic deformation; constructing pull-back measurement in the domain, solving three groups of harmonic functions, extracting contour surfaces according to common area constraint, and intersecting to form a network; deformation and harmonic functions are taken as joint unknown quantities, synchronous optimization is carried out in combination with pixel consistency, physical consistency and topological consistency, and the density of the inducer is re-weighted by layer area deviation feedback until convergence. According to the method, the surface net with consistent anatomy, controlled topology and balanced layer is output, and the method is adaptive to diagnosis, treatment and simulation application.

An optical measuring device (1) for the optical measurement of objects (O) comprising a motion platform (2) and a light line scanner is described. The light line scanner is connected to the motion platform (2) and is moved with the motion platform (2) into predefined measuring positions and orientations in order to project light lines (LL) from these measuring positions and orientations onto the object (O). The light line scanner is a multi-line scanner (3) designed to project several light lines (LL) onto the object (O).The method for the optical measurement of objects (O) using such an optical measuring device (1) involves moving the multi-line scanner (3) with the motion platform (2) successively to several different measurement positions and orientations relative to the object (O). In each measurement position and orientation, the corresponding axis positions of the axes of the motion platform (2) are measured, several light lines (LL) are projected onto the object (O), and images of the projections of the light lines (LL) projected onto the object (O) are captured in individual measurement images. From the measured axis positions of the axes of the motion platform (2), the respective position and orientation of the projections of light lines (LL) captured in the individual measurement images are determined.The individual measurement images are combined based on the respective determined positions and orientations to form a set of 3D points and / or a polygon mesh that describe the three-dimensional surface of the measured object (O).

A bitstream that includes attribute information of a plurality of attributes in a mesh is received. One or more prediction modes are determined from a plurality of candidate prediction modes for a current attribute of the plurality of attributes in the mesh based on one of prediction mode priorities and prediction mode accuracies of the one or more prediction modes in the plurality of candidate prediction modes. A prediction value of the current attribute in the mesh is determined based on the attribute information of the current attribute and the one or more prediction modes. The current attribute is reconstructed based on the prediction value of the current attribute.

And receiving a code stream, wherein the code stream comprises attribute information of at least two UV vertexes corresponding to the grid. A seam vertex queue is generated, the seam vertex queue including a seam vertex queue of at least one UV vertex of the at least two UV vertices. Each UV vertex of the at least one UV vertex in the seam vertex queue belongs to a corresponding UV vertex group of at least two seam UV vertex groups. Each of the at least two sets of seam UV vertices corresponds to a same three-dimensional (3D) vertex of the mesh in a three-dimensional (3D) space. An initial UV vertex of the mesh is determined based on whether one of the at least one UV vertex in the seam vertex queue includes an unaccessed UV vertex. And reconstructing UV coordinates of the grid based on the initial UV vertexes.

The invention provides a touch panel. The touch panel comprises a substrate, a first metal layer and a second metal layer, the first metal layer is arranged on the substrate and comprises a plurality of first polygonal meshes, wherein the length of each first polygonal mesh in the first direction is larger than the width of each first polygonal mesh in the second direction. The second metal layer is arranged on the substrate and comprises a plurality of second polygonal meshes, wherein the length of each second polygonal mesh in the second direction is larger than the width of each second polygonal mesh in the first direction.

This invention discloses a charging station information processing method, display method, device, system, and storage medium. The method includes: acquiring an initial map containing charging station information; generating a polygonal grid for the initial map, wherein each polygonal grid contains charging station information for at least one charging station; filtering the charging station information in the polygonal grid based on the correlation between preset point positions and the positions of charging stations in the polygonal grid to generate display data for each polygonal grid; and sending the display data to a vehicle-mounted system for generating a charging station display layer when the vehicle-mounted system displays the map. The technical solution of this invention minimizes data loading and reduces the performance consumption of the vehicle-mounted system while ensuring a good user experience.

The present invention belongs to the technical field of real-time grid mapping in outdoor unknown environments, and specifically is an incremental polygonal grid reconstruction method based on laser radar point cloud data. The method obtains and preprocesses point cloud data, uses point cloud data to perform voxel construction and incremental grid fusion, matches new point clouds to grids, performs incremental grid division of fused and registered point clouds, and performs polygonal grid simplified mapping steps, and finally obtains a polygonal grid map that can be used in subsequent mapping processes. The present invention improves the efficiency of grid construction in complex dynamic environments through incremental grid generation, and combines point-to-grid matching, grid simplification and storage optimization methods to solve the balance problem between real-time performance, accuracy and computing resources in robot system mapping.