Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

127 results about "Viewing cone" patented technology

When a visual display with non-vanishing size is seen by an observer, every point of the display area is seen from a different direction as illustrated in fig. 1. No two spots on the display are seen from the same direction. The larger the display is and the closer the observer is to the display the more the viewing direction varies over the surface area of the display.

Virtual reality building roaming method and system based on scene rendering

The invention provides a virtual reality building roaming method and system based on scene rendering, and relates to the technical field of virtual reality, and the method comprises the steps: dividing building three-dimensional model data through an octree space segmentation algorithm, and determining a detail level threshold according to a viewpoint distance, and constructing a multi-level detail model. And baking the basic illumination chartlet by using a pre-calculation irradiance transmission technology, and performing real-time synthesis in combination with the multi-resolution shadow chartlet. And calculating illumination by adopting a physics-based rendering method, and fusing the illumination with the scene light and shadow synthesis data to generate dynamic illumination rendering data. And determining a current viewpoint based on the spatial position and attitude data of the virtual reality head-mounted display, performing view cone elimination and shielding elimination, and performing GPU instantiation rendering on visible building data. And the rendered image is optimized through high dynamic range processing and a time domain anti-aliasing technology, and a final scene image is generated in combination with an atmospheric scattering algorithm. According to the invention, the rendering efficiency and scene reality sense of virtual reality building roaming can be effectively improved.
Owner:GUANGZHOU DABAI DIGITAL TECHNOLOGY CO LTD

Three-dimensional Gaussian sputtering optimization method for pose-free input

The invention discloses a three-dimensional Gaussian sputtering optimization method for pose-free input. The method comprises the following steps: for an input image, predicting light beam distribution by using a light prediction model to obtain distribution characteristics in a light beam form; calculating the pose of the camera based on the distribution characteristics of the light beam form; for light beam distribution, sampling is carried out based on the volume density of light, and initial space distribution of a three-dimensional Gaussian point cloud focused in a visual center area is obtained; for an input image, a visual shell is obtained through view cone projection and object mask calculation; three-dimensional Gaussian sputtering scene training is carried out based on the initial space distribution of the three-dimensional Gaussian point cloud and the visual shell, a three-dimensional scene reconstruction model meeting a set loss function standard is obtained, and a loss function comprises training regular terms of camera pose parameters. According to the method, important initialization scene information is provided for three-dimensional Gaussian sputtering training, and the quality and detail richness of a final three-dimensional structure are remarkably improved.
Owner:SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI

Image rendering method and device, electronic equipment and storage medium

The invention provides an image rendering method and device, electronic equipment and a storage medium, and the method comprises the steps: calculating a distance between a virtual image obtaining device and a to-be-rendered object through determining the to-be-rendered object in a to-be-rendered scene, and determining the distance as a first loading factor; determining a rendering picture pixel proportion of each to-be-rendered object based on preset view cone size information, determining the rendering picture pixel proportion as a second loading factor, generating a rendering loading priority sequence according to the first loading factor, and determining a model level resource file and a texture level resource file according to the second loading factor, loading the resources to a video memory for rendering according to the sequence; the priority is determined by calculating the distance between the camera and the to-be-rendered object, it is ensured that a high-quality model and texture of a close-range object are loaded preferentially, the instantaneity of the rendering loading process is improved, texture and model grade division is conducted through the pixel proportion of the to-be-rendered object in the picture, and the rendering efficiency is improved. The occupation of a video memory is reduced, and the overall rendering performance of the system is improved.
Owner:CISDI ENGINEERING CO LTD +1

Three-dimensional scene optimization method and system for collaborative rendering of dynamic LOD and view cone elimination based on space-time prediction

The invention relates to the field of scene rendering, and particularly discloses a three-dimensional scene optimization method and system for collaborative rendering of dynamic LOD and view cone rejection based on space-time prediction, and the method comprises the steps: dynamically calculating the visibility frequency of an object, and dynamically adjusting the loading and rendering modes of objects with different priorities; predicting a view cone range of multiple frames in the future by using an LSTM network architecture; dividing the scene into uniform grid blocks, and then performing grading elimination; optimizing a heterogeneous computing pipeline; rendering the execution process; the system comprises a dynamic LOD and visual cone rejection collaborative optimization module, an LSTM visual cone prediction module, a block-level mixed rejection module, a heterogeneous calculation pipeline module, an optimization CPU-GPU task allocation and data transmission module and a rendering execution control module. According to the method, the technologies of collaborative optimization of dynamic LOD and view cone removal, block-level mixed removal, heterogeneous calculation assembly line optimization and the like are adopted, so that unnecessary rendering calculation and resource loading are reduced, and the rendering efficiency is improved.
Owner:YANTAI JIERUI NETWORK TRADING

OpenGL-based large-scale three-dimensional model object rapid pickup method

The invention relates to an OpenGL (Open Graphics Library)-based large-scale three-dimensional model object quick pickup method. The method comprises the following steps: converting an interactive operation of a user into a three-dimensional space structure; a CPU constructs octree space acceleration structures based on a bounding box of an object in a large-scale three-dimensional model, and leaf nodes, intersecting with a pickup ray or a pickup view pyramid, of an axis alignment bounding box in each octree space acceleration structure are screened out to serve as candidate leaf nodes; collecting triangles contained in all candidate leaf nodes, and constructing a candidate triangle set; dynamically allocating thread groups and threads of the GPU based on the number of candidate triangles in the candidate triangle set; and performing highlight display on the candidate triangles intersected with the pickup ray or the pickup view cone, and feeding back IDs of the corresponding candidate triangles. According to the method, the data transmission quantity and the video memory access pressure are remarkably reduced, the pickup response time is greatly shortened in a large-scale three-dimensional model scene, and the requirements of real-time interaction and visual operation can be met.
Owner:SHENZHEN MAIXI SOFTWARE CO LTD

Camera external parameter calibration method and device based on image and point cloud matching

The invention provides a camera external parameter calibration method and device based on image and point cloud matching, and the technical scheme of the invention is that a point cloud rendering view corresponding to an original point cloud is rotated, so that the point cloud rendering view and a camera image are superposed visually, and initial space association is established; and directly learning and solving accurate camera external parameters from the local relevance between the original point cloud intensity information and the image RGB information in an end-to-end manner by using a deep learning model. According to the scheme, the strict camera-radar orientation consistency requirement in a traditional method is not needed through rough matching, manual participation in the whole process of traditional calibration is replaced, transition from visual alignment to geometric alignment is achieved, and the calibration efficiency and scene adaptability are remarkably improved. Meanwhile, the scheme of the invention initiates a process of visual field cone cutting-perspective projection rasterization-intensity-RGB end-to-end matching, and compared with a traditional feature point method, the calculation complexity is greatly reduced, and the memory occupation is reduced by 80%.
Owner:BEIJING GREEN VALLEY TECH CO LTD +3

Virtual shooting method and device

The invention relates to a virtual shooting method and device, which are applied to virtual shooting performed by adopting a plurality of shooting devices, and the method comprises the following steps: in a first display period, displaying a plurality of inner view cone pictures on a display screen in a time-sharing manner; wherein any shooting device in the plurality of shooting devices corresponds to at least one inner view cone picture in the plurality of inner view cone pictures, and different shooting devices correspond to different inner view cone pictures; the time-sharing display means that an inner view cone picture corresponding to any shooting device is displayed in a first time slice corresponding to the shooting device in the first display period, and inner view cone pictures corresponding to different shooting devices are displayed in different first time slices; and shooting in the corresponding first time slice through each shooting device in the plurality of shooting devices, so that each shooting device shoots a corresponding inner view cone picture. According to the invention, multi-camera virtual shooting is realized, and the manufacturing efficiency and the application scene of virtual shooting are greatly improved.
Owner:DIVINE VISION (SHENZHEN) CULTURE TECH CO LTD

Multi-precision dynamic loading and performance optimization rendering method for hydroelectric generating set model

The invention discloses a multi-precision dynamic loading and performance optimization rendering method for a hydroelectric generating set model. The method comprises three stages of multi-precision model construction, multi-precision model dynamic loading and rejection, and adaptive rendering optimization. And a multi-precision model construction stage: performing component extraction on the original model file to generate high, medium and low resolution multi-precision models, and converting the models into a GLTF format to improve the loading efficiency. And a dynamic loading and rejecting stage: reading the multi-precision GLTF file, generating a geometric model by applying the conversion matrix, constructing an LOD model according to the sight distance, and rejecting only the visible part in the loading visual field through a view cone to reduce the rendering calculation amount. And an adaptive rendering optimization stage: monitoring the frame rate in real time, dynamically adjusting the resolution when the frame rate is relatively low, simplifying material illumination, forbidding special effects such as shadow and the like, ensuring stable rendering performance, and finally outputting an optimized three-dimensional scene. According to the method, the rendering speed and fluency of the water turbine model are remarkably improved, and the method is suitable for the real-time visualization requirement of a complex three-dimensional scene.
Owner:HOHAI UNIV +1

Terrain rendering method and device for virtual scene, equipment, medium and program product

The invention provides a terrain rendering method and device for a virtual scene, electronic equipment, a computer readable storage medium and a computer program product, and the method comprises the steps: obtaining a plurality of three-dimensional grids in a view cone corresponding to the virtual scene, and enabling the three-dimensional grids to be obtained through dividing a three-dimensional terrain in the virtual scene; for each three-dimensional grid, obtaining a division grade corresponding to the three-dimensional grid, and determining a texture grade of the three-dimensional grid based on the division grade; obtaining texture parameters of the three-dimensional grids in the rendering process based on the texture levels of the three-dimensional grids, and rendering the plurality of three-dimensional grids based on the texture parameters and the division levels corresponding to the three-dimensional grids to obtain a target image; wherein the target image is used for indicating a topographic image in a view field range indicated by the view cone in the virtual scene. In this way, the rendering efficiency and the rendering effect can be improved when the terrain of the virtual scene is rendered.
Owner:TENCENT TECHNOLOGY (SHENZHEN) CO LTD

Game rendering method and system capable of eliminating volume cloud and mist in real time and electronic equipment

The invention provides a game rendering method and system capable of eliminating volume cloud and mist in real time and electronic equipment, and belongs to the technical field of game development. The method comprises the steps that a cloud and mist area is set, the set cloud and mist area is divided into a plurality of equal-size grids, and each grid stores an elimination state and an elimination progress value; dynamically updating vertex concentration textures; calculating and storing cloud and mist concentration values of vertexes around each grid based on the grid elimination progress; view cone optimization rendering: determining a visible cloud range, combining adjacent grids to be rendered into a rectangular region in a two-dimensional plane, expanding the rectangular region into a cube, and projecting the cube to a screen space to calculate a bounding box; and shader sampling: for pixels in the screen bounding box, calculating cloud transparency and color through ray stepping sampling. According to the method, through grid dynamic management, rendering range optimization and shader efficient sampling, low-consumption and real-time elimination high-quality volume cloud and mist rendering can be realized at a mobile terminal, and performance and game interaction experience are considered.
Owner:GUANGZHOU AIYOU INFORMATION TECH

Method and system for removing invisible objects based on GPU (Graphics Processing Unit)

The invention discloses an invisible object removing method and system based on a GPU, and the method comprises the steps: calculating six plane equations of a view cone of an observation camera and the world coordinates of each observed object, and carrying out the parallel calculation of the front or back of each observed object on a certain plane of the view cone through the GPU, thereby judging whether the object is outside the view cone or not, therefore, all objects outside the view cone are removed. And generating hierarchical Z buffer area data in the GPU through the depth map of the current frame, transmitting information such as the position of the object to the GPU for calculation with the hierarchical Z buffer area data, and removing the completely shielded object. According to the method, the parallel computing capability of the GPU is utilized, the rejection efficiency is improved, the CPU computing burden during rendering is greatly reduced, and the occurrence probability of rejection errors is reduced.
Owner:ZHEJIANG UNIV

Engineering BIM model lightweight rendering method based on model semantic information

The invention provides an engineering BIM model lightweight rendering method based on model semantic information, and the method comprises the steps: analyzing an engineering BIM model, and obtaining the semantic information of the engineering BIM model; performing octree space segmentation on the engineering BIM model, and establishing a model octree optimization structure of the engineering BIM model; semantic information of each tile of each level is obtained; and carrying out space elimination based on view cones and semantic elimination based on model semantics. By means of the method, the efficiency of octree index calculation and model rendering simplified calculation of the engineering BIM model can be improved, the data size of the BIM model rendered by a browser at the same time is reduced, the utilization rate of hardware such as a GPU is increased, the loading and rendering efficiency of the BIM model is improved, more BIM models can be loaded at the same time under the same hardware configuration condition, and the loading efficiency and the rendering efficiency of the BIM model are improved. And rendering application of massive BIM models is supported.
Owner:POWERCHINA BEIJING ENG CORP

Anode assembly workshop three-dimensional visual management system and method based on digital twinning

The invention discloses a digital twinning-based three-dimensional visual management system and a digital twinning-based three-dimensional visual management method for an anode assembly workshop. The system comprises a triple binding index module, a unified time axis alignment module, a view cone inverse solution positioning module, a safety linkage arrangement module and a resume playback module, and is provided with a process template library and a network isolation / main / standby module. Three-dimensional positioning and PTZ linkage of an AI / threshold event are realized by establishing mapping of a three-dimensional object, a camera and a PLC point location; the batch / bracket / station events, the PLC time sequence and the video slices are indexed in a unified mode; shadow verification, interlocking check, instruction issuing, read-back and video consistency confirmation and trace leaving are executed according to the process, and a safe closed loop from warning to disposal is formed. The method comprises the steps of access and calibration, triple binding, time axis alignment, event positioning and stream taking, linkage processing and resume filing. The positioning and handling efficiency can be improved, the misoperation risk is reduced, and cross-production-line reuse and tracing evidence obtaining are supported.
Owner:QINGTONGXIA ALUMINUM GRP

Three-dimensional target detection method based on cone point cloud clustering and 2D detection

The invention discloses a three-dimensional target detection method based on view cone point cloud clustering and 2D detection, and the method comprises the steps: generating a plurality of target bounding boxes with categories according to a 2D image, converting the target bounding boxes to an image coordinate system, and carrying out the screening of 3D point clouds, so as to obtain view cone point clouds; performing a point cloud clustering operation on the view cone point cloud, and calculating a result of the point cloud clustering operation to obtain a 3D frame meeting a preset requirement; and performing IOU calculation of different visual angles on the 3D frames to merge the 3D frames of the same object under different visual angles so as to obtain a 3D bounding box. According to the method, the problem of geometric information loss can be effectively avoided, and the accuracy of the finally obtained 3D bounding box is ensured; the problem that global search is needed due to the fact that 3D point cloud is directly adopted is avoided, the complexity of the calculation process is reduced, and the overall calculation efficiency is effectively improved; the preset frame requirement can be set according to the requirement of a use scene, so that different requirements are met, and the wide use range is ensured.
Owner:城市之光(深圳)无人驾驶有限公司

Multi-modal depth fusion 3D target detection method based on reliable depth estimation

The invention discloses a multi-modal depth fusion 3D target detection method based on reliable depth estimation, and the method comprises the steps: firstly constructing an image feature extraction network and a millimeter-wave radar point cloud feature extraction network, and respectively extracting image features and millimeter-wave radar point cloud features; secondly, fusing the image and millimeter wave radar point cloud features to obtain depth probability distribution, multiplying the depth probability distribution by the image features to obtain image view cone features, making a depth probability distribution label by using the laser radar point cloud after depth completion, and performing supervision training based on the depth probability distribution; then combining the view cone features of the image and the point cloud features of the millimeter wave radar, and obtaining a feature FA by using a Splat network and a deformable attention mechanism; and finally, based on the feature FA, completing 3D target detection by using a 3D target detection head, and outputting a detection result. According to the method, depth estimation is optimized through the millimeter wave radar, so that the accuracy and robustness of 3D target detection are improved.
Owner:HANGZHOU DIANZI UNIV

Marine terrain digital display method based on three-dimensional modeling

The invention discloses an ocean terrain digital display method based on three-dimensional modeling. The ocean terrain digital display method comprises the following steps of 1, constructing a multi-source ocean data set; 2, performing space fusion by adopting a self-adaptive interpolation method based on position weight to generate a three-dimensional space density field; 3, performing contour surface extraction by adopting an improved Marching Cubes algorithm to generate an initial three-dimensional terrain model; 4, performing local modeling by adopting a Delaunay triangulation network reconstruction method, and constructing a complete three-dimensional terrain model; 5, obtaining a texture terrain model; step 6, constructing an LOD tree structure and realizing dynamic loading in combination with view cone cutting and a screen space error control strategy; and step 7, based on a dynamic loading result of the LOD tree structure, driving node switching through a time controller, and performing visual display. According to the method, three-dimensional modeling and an LOD tree structure are fused, and high-precision reconstruction and evolution display of the ocean topography are achieved.
Owner:XIAMEN OCEAN VOCATIONAL & TECH COLLEGE

Mechanical arm fine positioning and closed-loop calibration method and system based on heterogeneous visual fusion

The invention discloses a mechanical arm fine positioning and closed-loop calibration method and system based on heterogeneous visual fusion, and the method comprises the steps: heterogeneous data synchronous collection and cross-modal joint calibration, semantic segmentation based on prompt learning and high-precision mask generation. Point cloud refining and edge noise elimination in a view cone space; and differential compensation and closed-loop feedback control based on a golden model. A heterogeneous visual fusion framework is constructed, and 3D point cloud noise is constrained by using a sub-pixel edge of a high-resolution 2D image, so that the positioning precision breaks through a sub-millimeter level; a segmentation large model with zero sample generalization ability and a prompt learning mechanism are introduced, so that the system can adapt to various workpieces under a complex background without repeated training; refining point cloud and a standard three-dimensional gold model are registered and fed back to the mechanical arm in a closed loop mode, so that the system can compensate and calibrate drift and accumulative errors in real time, and long-term stable operation is achieved.
Owner:NINGBO XINGBOYUAN INTELLIGENT TECHNOLOGY CO LTD

Shielding object removing method and device, electronic equipment and storage medium

PendingCN119488716AVideo gamesViewing coneVisible object
The embodiment of the invention provides an occluded object removing method and device, electronic equipment and a storage medium, and relates to the technical field of three-dimensional images. The method comprises the steps that a building group structure of a virtual scene is established according to the virtual scene to be rendered, the building group structure comprises a room, a connection entrance and a shielding body, and the shielding body represents an object which may be invisible during image rendering; performing view cone cutting according to the current position of the player in the virtual scene and the connection entrance, and calculating a cut view cone body of the current position; and calculating a shielding body which is not intersected with the cutting view cone in each room as a target shielding body, and removing each target shielding body. Therefore, through the constructed room, the connection entrance and the shielding body, whether the object is shielded or not can be rapidly calculated during operation, and a large number of invisible objects can be rapidly removed.
Owner:BEIJING PIXEL SOFTWARE TECH

3D point cloud segmentation method and device, electronic equipment and storage medium

The invention provides a 3D point cloud segmentation method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining 3D point cloud data and image data corresponding to the same sampling moment and in the same scene; converting the image data into a set of 2D masks, each 2D mask identifying a contour region of an object contained in the image data; the pixel coordinates of each 2D mask are converted into a 3D point cloud coordinate system, so that the corresponding view cone space of each 2D mask in the 3D point cloud coordinate system is determined; and for each 2D mask, segmenting 3D point cloud data in a view cone space corresponding to the mask from the 3D point cloud data to determine a point cloud block where the object is located. Therefore, 3D point cloud data does not need to be manually labeled, point cloud segmentation can be guided only by using the mask generated by the image, and the manual labeling cost is reduced; the method does not depend on specific sensor parameters, adapts to different hardware configurations, reduces the adaptation cost of different platforms, and enhances the generalization ability.
Owner:ZHEJIANG GEELY HLDG GRP CO LTD +1

Scene data scheduling and rendering method based on dual PVS

The invention discloses a scene data scheduling and rendering method based on dual PVS, and relates to the technical field of data processing. According to the method, the target scene unit where the camera viewpoint is located is determined, the rendering scene tree corresponding to the target scene unit is obtained, the nodes of the rendering scene tree are the target visible objects corresponding to the target scene unit, and then the view cone corresponding to the camera viewpoint is determined; and selecting a target node in the rendering scene tree as a target rendering object according to the view cone, determining a hierarchical detail model corresponding to the target rendering object according to the distance between the target rendering object and the camera viewpoint, loading the hierarchical detail model to the rendering pipeline from the rendering scene tree, and rendering the target rendering object. According to the method, the target visible objects are screened through the rendering scene tree, and then the target rendering object is determined in the target visible objects through the view cone, so that the objects needing to be rendered are further reduced, and the computing resource consumption of model rendering is reduced.
Owner:SHENZHEN EMAP INFORMATION CO LTD

Road tunnel environment traffic condition monitoring method and device

The invention discloses a road tunnel environment traffic condition monitoring method and device. The method comprises the steps of obtaining image data and point cloud image data; extracting image features from the image data, screening feature points of the point cloud image data according to curvature, and matching feature points of front and back frames; combining the image features with depth information in the matched point cloud image data, and performing depth estimation, view cone representation and pooling operation on the combined features to obtain image BEV features; local feature extraction and maximum pooling operation are carried out on point cloud data in the point cloud image data, and after image features are fused, laser radar BEV feature prediction is obtained; and constructing a world coordinate system, and in the world coordinate system, carrying out cascading, feature extraction, global average pooling and convolution prediction on image BEV feature prediction and laser radar BEV feature prediction to obtain dynamic fusion features. According to the scheme provided by the invention, effective fusion of the laser radar point cloud data and the focal length camera image data can be realized.
Owner:HUNAN COMM RES INST CO LTD

Visual landslide monitoring method and device

The invention relates to a landslide visual monitoring method and device, and belongs to the technical field of landslide monitoring, and the method comprises the steps: obtaining the disaster scene data of a landslide monitoring area, constructing a view cone of the landslide monitoring area, dividing the disaster scene data through a plurality of LOD layers of the view cone, and obtaining the landslide scene data of the landslide monitoring area. Initial disaster scene data of each LOD level are obtained, error calculation and judgment can be carried out on the initial disaster scene data of each LOD level, and therefore multi-level visual analysis of the disaster scene data can be achieved; and the initial disaster scene data of each target LOD level can be subjected to adaptive scheduling and visual display, so that the intelligent degree and the adaptive capability of the disaster scene are improved, and the visualization efficiency is improved.
Owner:CCCC SECOND HIGHWAY CONSULTANTS CO LTD

Three-dimensional Gaussian point cloud rendering method based on graph pipeline

The invention discloses a three-dimensional Gaussian point cloud rendering method based on a graphic pipeline. According to the method, the calculation shader is used for point cloud preprocessing and attribute calculation, the graphic pipeline is used for hardware rasterization and color mixing, the characteristics of the GPU are utilized to the maximum extent, and the extremely high rendering speed is achieved while the universality is higher. The method comprises the following stages: 1, a point cloud preprocessing stage: calculating point cloud attributes and NDC coordinates, and removing view cones at the same time; 2, performing depth sorting; and 3, coloring the point cloud. And 4, a graph pipeline stage, including point cloud Gaussian splashing rasterization and color transparency mixing. Compared with original CUDA rendering, the method can be operated on any platform supporting the graphic pipeline in the environment of a NVIDIA display card which does not need to be operated, higher universality is achieved, meanwhile, rasterization of point cloud Gaussian splashing is conducted through the graphic pipeline, the rasterization and color mixing can be conducted through the graphic pipeline, hardware acceleration can be naturally used, and the method has the advantages of being high in practicability and easy to popularize. Therefore, the extremely high rendering speed is realized.
Owner:HANGZHOU DIANZI UNIV

Water conservancy twin model rendering optimization method for dynamically generating grayscale image frames based on view cone

The invention provides a water conservancy twinborn model rendering optimization method for dynamically generating gray level image frames based on a view cone. The method comprises the following steps: obtaining visible area model data after the current view cone is dynamically cut; outputting the visible area model data to a multi-layer rendering target comprising a color buffer area, a vertex buffer area and a depth buffer area through off-screen rendering so as to generate intermediate frame data isolated from screen display; performing gridding processing on the intermediate frame data, and displaying the equipment resolution according to the physical range size of the water conservancy project or the drainage basin scene; the average color value of each grid unit is calculated, desaturation processing is carried out to generate a basic grey-scale map, and adaptive feature enhancement processing is carried out to generate an optimized grey-scale map or a gradient map; and a GPU rendering pipeline is input and acts on hydrological feature analysis of the post-processing stage of the current frame and a preloading data source rendered by a twinborn model of the next frame at the same time, historical frame data are reused in a static region through motion detection, and real-time rendering is triggered in a dynamic region.
Owner:FUJIAN ZHONGRUI HANDING DIGITAL TECH CO LTD

Ultrahigh-resolution immersive scene rendering method based on hexagonal picture segmentation

The invention discloses an ultrahigh-resolution immersive scene rendering method based on hexagonal picture segmentation. The method comprises the following steps: 1, determining a hexagonal segmentation style of an immersive display surface; 2, unfolding the immersive display surface into a plane; 3, acquiring 3D point information of each hexagonal surface, respectively calculating the width W and the height H of the enveloping quadrilateral surface, and calculating a local coordinate system V of the hexagonal surface S; 4, using W, H and V to construct a view cone and a viewpoint matrix of the three-dimensional scene; 5, setting a shade M of a plane expansion view for the hexagonal surface S; 6, calling a rendering scene of the 3D scene for the mask region by using information of the view cone and the viewpoint matrix, and rendering the mask region into a picture at the position S of the hexagonal surface; and 7, repeating the steps 5-6, and sequentially carrying out the same treatment on each hexagonal surface. According to the method and the device, the ultrahigh-resolution picture can be rendered in a scene segmentation form, and the method and the device are suitable for ultrahigh-resolution picture rendering under immersive display equipment.
Owner:HARBIN AIWELL TECH CO LTD

Pure vision automatic driving environment sensing method and system based on variable depth mechanism

The invention discloses a pure vision automatic driving environment perception method and system based on a variable depth mechanism, and the method comprises the steps: obtaining original images of different visual angles around a driving subject, and carrying out the processing of the original images of different visual angles, and obtaining a multi-visual-angle feature; performing difference processing on the multi-view feature at the current moment and the multi-view feature at the previous moment to obtain a time sequence difference feature containing time sequence information; performing depth information extraction on the time sequence difference features by using a variable depth mechanism to obtain variable depth distribution; performing outer product operation on the variable depth distribution in the time sequence difference features to generate a view cone feature matrix; mapping the view cone feature matrix to a BEV space with the vehicle as the center, and generating an environment perception result; according to the invention, the accuracy of object depth estimation in image-based automatic driving environment perception is improved.
Owner:HEFEI INSTITUTE OF PHYSICAL SCIENCE CHINESE ACADEMY OF SCIENCES

Image BEV visual angle conversion method based on negative probability and depth estimation

The invention relates to an image processing technology, and discloses an image BEV visual angle conversion method based on negative probability and depth estimation, which comprises the following steps: extracting a semantic feature image set of a forward view image under each visual angle; performing depth estimation and probability mapping on the front view image semantic feature image set to obtain initial probability distribution, and summarizing the initial probability distribution to obtain an initial probability distribution matrix; performing negative probability transformation on the initial probability distribution matrix to obtain a negative probability distribution matrix; performing feature fusion on the negative probability distribution matrix and the initial probability distribution matrix with the front view image semantic feature image set to obtain plane features; and converting the plane features into view cone features and mapping the view cone features to a three-dimensional space to obtain a BEV view angle feature map. The invention also provides an image BEV visual angle conversion device and equipment based on negative probability and depth estimation, and a medium. According to the invention, the accuracy of image BEV visual angle conversion based on negative probability and depth estimation can be improved.
Owner:CHONGQING UNIV

Roadside multi-mode fusion sensing method

The invention discloses a roadside multi-mode fusion sensing method. According to the method, an initial 3D view cone tensor with depth information is generated through a depth estimation network, features of pixel points are projected to a coordinate system in combination with internal and external parameter matrixes of a camera, then a three-dimensional space point coordinate set is generated according to projected feature position information, a Gaussian function is adopted to distribute weights of neighborhood grids, and a three-dimensional space point coordinate set is obtained. And finally, inputting the image features and the features into a fusion unit, respectively carrying out uncertainty modeling on the two types of features, automatically generating a fusion weight, and finally, carrying out normalization weighted aggregation on the features in the neighborhood so as to reduce the position approximation error and obtain the calibrated image features, and finally, inputting the image features and the features into the fusion unit, and respectively carrying out uncertainty modeling on the two types of features and automatically generating a fusion weight. And dynamic optimization is carried out according to a self-supervised loss function in the fusion process, so that the stability of multi-modal feature fusion is ensured, the detection accuracy is high, the performance is relatively good, and the problem of global feature dislocation caused by external parameter deviation and depth error can be avoided.
Owner:NANJING UNIV OF SCI & TECH

Obstacle detection method and device, fusion processing method and device, electronic equipment and storage medium

The invention discloses an obstacle detection method and device, a fusion processing method and device, electronic equipment and a storage medium, and the method comprises the steps: responding to 3D laser point cloud data of a laser radar, and restoring a corresponding visual cone of a target in a 3D world according to a 2D detection frame of a camera and camera parameters; a pre-trained target detection model is adopted to process the point cloud data of the target in a corresponding view cone in the 3D world, and target point cloud data of the laser radar in the view cone at the same moment are obtained; and according to the target point cloud data, processing the target point cloud data through the target detection model to obtain an obstacle detection result. The foreground and background point cloud distinguishing capability of the target detection model is improved.
Owner:MUSHROOM CHELIAN INFORMATION TECH CO LTD

Lightweight rendering method for webgl-based digital twin model of pipe network

The present application relates to a kind of WebGL-based pipe network digital twin model light weight rendering method, method includes: resource preconfiguration: the grid data of different three-dimensional model is generated corresponding instance template;Through instance template generation instance is stored in instance pool;In scene graph, create include call data and dynamic data lightweight placeholder, call data is used to extract instance, and dynamic data is used to adjust the dynamic display of instance;Lightweight rendering: set the visibility screening strategy of placeholder, whether current placeholder is visible is judged in turn through the screening condition of level, distance and perspective, view cone intersection;Set instance management strategy, for the placeholder screened as visible, corresponding instance is called from instance pool and is bound, and the binding relationship is recorded, for the placeholder screened as invisible, unbind, and the instance after unbinding is marked as idle;The placeholder of bound instance is rendered and output.Effectively reduce the memory occupation when rendering.
Owner:ZHONGRUIHENG (BEIJING) TECH CO LTD