19 results about "Point mapping" patented technology

This invention discloses a training-free subjective-objective video object mapping method driven by spatial feature alignment. First, videos with target object annotations in both subjective and objective perspectives are designated as source perspective videos, while the other perspective video is designated as the target perspective video. A pre-trained dense feature matching model is used to process the source and target perspective videos, obtaining a bidirectional feature point mapping relationship and a matching confidence map. Based on the target object mask of the source perspective video, an effective feature point set is obtained. Based on this effective feature point set, keyframes and the target object masks of the corresponding frames in the target perspective video are selected and used as spatiotemporal cues. A pre-set cue-based video object segmentation model is then used for target segmentation, yielding the target object mask of the target perspective video frame, thus obtaining the subjective-objective video object mapping relationship. This invention requires no training and significantly improves the accuracy of subjective-objective video object mapping tasks on large-scale datasets by utilizing feature matching and cue-based video object segmentation.

The present application relates to teaching robot practice evaluation related technology, provide a kind of teaching robot operation trajectory three-dimensional evaluation method, system, equipment and storage medium.The method obtains the image sequence of teaching robot in the process of executing practice task collected by fixed external camera, identifies and continuously tracks end effector, tool center point or preset key position, obtains two-dimensional operation trajectory information;According to the camera internal parameter and external parameter, the two-dimensional trajectory point is mapped into space ray, and the three-dimensional operation trajectory under world coordinate system is solved in combination with constraint optimization model;Again, according to key event, key action node or preset space region, action stage is divided and time sequence is aligned, and matching score is generated for each action stage and output trajectory score result.The scheme is suitable for practice task process evaluation in teaching robot platform.

This invention belongs to the field of thermal power unit automation control technology, and discloses a configuration batch replacement method and related device based on hash caching technology. The configuration batch replacement method includes: reading a modified measurement point mapping table to obtain all source KKS and target KKS; updating the original hash table cached in memory based on the obtained source KKS and target KKS to obtain an updated hash table; traversing the updated hash table to perform configuration batch replacement; wherein, during configuration replacement, location information is obtained through the measurement point type, and then the source KKS in the indexed variable is replaced with the target KKS through the location information index variable. This invention can achieve fast and accurate replacement of KKS within the configuration, helping engineers to perform efficient configuration design; this invention solves the technical problems of slow configuration replacement speed and high system resource consumption in existing technologies, and has high engineering application value.

This application relates to the field of geographic information technology and discloses a method for 3D display of pipeline inspection videos, comprising: acquiring a 3D model of a target pipeline and an inspection video of the interior of the target pipeline; correcting the pipeline center point and pipeline cross-section for each frame of the inspection video to obtain corrected video frame images; acquiring a panoramic image of the interior of the target pipeline based on each corrected video frame image; mapping the pixels in the panoramic image to the 3D model of the pipeline to generate a textured 3D model of the pipeline; and displaying the textured 3D model of the pipeline in a preset 3D scene. This method improves the accuracy of displaying the internal condition of the pipeline in a 3D scene. This application also discloses an apparatus, electronic device, and storage medium for 3D display of pipeline inspection videos.

This application provides a method, apparatus, device, and storage medium for controlling the display of a 3D digital human. The method, applied to a 3D display device, includes: acquiring an image sequence of a digital human; wherein the image sequence is generated based on a phoneme sequence of a response speech to a target user's dialogue; mapping each facial image in the image sequence to the screen of the 3D display device based on a multi-view rendering table corresponding to the digital human, so as to adjust the displayed digital human when the response speech is played; the multi-view rendering table corresponding to the digital human is used to describe the pixels in the facial images mapped to each point on the screen. This enables the display of a dynamic, 3D digital human during dialogue, based on a multi-view rendering table, improving the stereoscopic effect of the digital human and enhancing the user's dialogue experience.

The application discloses a kind of two-dimensional space to three-dimensional space key point mapping conversion method and system, including sequence preprocessing module: the length of two-dimensional space key point sequence is obtained, carries out sparse preprocessing, obtains sparse two-dimensional space key point sequence;Mapping dimension-increasing module: sequence modeling is carried out to sparse two-dimensional space key point sequence, obtains three-dimensional space key point sequence after preliminary mapping;Sequence aggregation module: three-dimensional space key point of center target is obtained by compressing and aggregating three-dimensional space key point sequence after preliminary mapping;Training network module: end-to-end training is carried out to network using supervised learning, and mapping accuracy of three-dimensional space key point estimation network is improved using space-time constraint strategy;Visualization module is used to obtain visual result.The application has good robustness and universality, and can be widely applied in key point detection of a variety of objects.

The application discloses a world time continuous solution method, comprising: obtaining and preprocessing the solution result files of each session in a predetermined time range; applying uniform constraints to each session after preprocessing, and performing parameter blocking and local parameter elimination; constructing a sliding time window in a target time range, parameterizing the world time by using a segmented linear function in the window, and establishing a session and time node mapping relationship; constructing a sliding time window on a global time axis, collecting sessions in the window, superimposing a session weight equation, adding a time node smoothing constraint, and jointly solving the world time residual increment; superimposing a priori on the joint solution result, recovering a tidal term, and generating a world time continuous sequence. The application solves the problems of the existing world time product, such as limited time resolution based on daily parameter level combination, insufficient utilization of session level observation geometry and equation information, sensitivity to modeling differences of different session types, and difficulty in adapting to the existing reference framework.

This invention relates to the field of industrial control technology and discloses a simulation platform and method for a pipeline control system based on isomorphic simulation of a real environment. The method includes: acquiring model parameter data, control engineering data, and point configuration data; establishing a pipeline physical model and a virtual programmable logic controller; establishing a point mapping table based on the point configuration data and generating point semantic fingerprints for the mapped points; performing collaborative simulation scheduling, writing the state variables of the pipeline physical model into the input image area according to the point mapping table, while keeping the pump unit control variables and valve control variables corresponding to the output image area unchanged; dividing the process into multiple sub-cycles and updating the pipeline physical model according to the sub-cycles; and performing shadow consistency verification, extracting the real point semantic fingerprint and comparing it with the point semantic fingerprint, outputting a zero-modification deployment package when consistency is satisfied. This application achieves verifiable consistency with zero-modification deployment, reducing deployment risk and shortening the joint debugging cycle.

The application relates to the field of medical image data processing and simulation systems, and discloses a tumor puncture real-time simulation system based on VR glasses, which comprises an image data preprocessing module, a real-time simulation module and a rendering pipeline module.The image data preprocessing module generates a physical particle cloud, a skin rendering grid and determines the physical constraint stiffness parameters between particles based on the gradient amplitude value of medical images.The real-time simulation module only interacts with the physical particle cloud.The rendering pipeline module passively drives the deformation of the skin rendering grid.The application objectively translates the mechanical properties by establishing the isomorphism between the image space relationship and the physical constraint relationship, and avoids the problems of traditional point-to-point mapping.

A system including a processing circuit configured to receive tags describing points of a piece of equipment, the piece of equipment connected to the system. The processing circuit configured to map the tags to classes of a schema of a graph data structure. The processing circuit configured to perform clustering to generate clusters of the points. The processing circuit configured to identify, based on the clusters, relationships in the schema of the graph data structure between the tags mapped to the classes of the schema of the graph data structure. The processing circuit configured to communicate data to a second system based at least in part on the tags mapped to the classes and the relationships.

This application discloses a method, apparatus, device, and storage medium for generating a 3D model, relating to the field of computer graphics technology. The method includes: sampling P mapping points from each pixel of a texture image; mapping the P pixels into a 3D space to obtain P mapping points in the 3D space; determining the number of meshes assigned to each of the P mapping points, wherein at least two of the P mapping points have different numbers of meshes assigned; and establishing a connection relationship between the P mapping points based on the number of meshes assigned to each of the P mapping points to generate a 3D model. This method reduces the complexity of the 3D model while maintaining its detailed representation.

Embodiments are disclosed for neural texture mapping. In some embodiments, a method of neural texture mapping includes obtaining a plurality of images of an object, determining volumetric representation of a scene of the object using a first neural network, mapping 3D points of the scene to a 2D texture space using a second neural network, and determining radiance values for each 2D point in the 2D texture space from a plurality of viewpoints using a second neural network to generate a 3D appearance representation of the object.

The application provides a 3DGS scene modeling and map management method and equipment for autonomous navigation, a unified voxel corner point space index benchmark is established, discrete perception anchor points are generated based on multi-modal sensor data, the generation and update of three-dimensional Gaussian continuous representation are driven, and the decoupling of rendering attributes and geometric structure is realized; the perception anchor points are mapped to generate corner point navigation units by using the unified index, and through the whole life cycle management and access control mechanism, only mature and stable navigation units are retained to participate in the pose constraint and map output, and dynamic noise is effectively eliminated; when the map is expanded, the continuity of the navigation constraint is maintained through the structure inheritance and gradual switching strategy across the sub-maps. The application converts the unstructured Gaussian radiation field into a structured navigation map which can be directly used for path planning, solves the problems that the existing 3DGS map is difficult to directly navigate and unstable in a dynamic environment, and effectively improves the precision and robustness of the autonomous navigation of mobile robots in a large-scale scene.

This application relates to the fields of artificial intelligence and insurance underwriting technology, proposing a dynamic interactive guidance method, device, equipment, and medium for reconstructing user health status. It can be applied to fintech scenarios. The method includes: constructing a multimodal material library for health scenarios; building an end-to-end feature reasoning architecture to perform cross-modal feature fusion of static virtual avatars and dynamic driving materials, outputting feature point mapping relationships and semantic consistency constraint parameters; based on the feature point mapping relationships and semantic consistency constraint parameters, performing action transfer and image reconstruction to generate a dynamic virtual avatar sequence adapted to the health scenario; precisely controlling the local actions of the dynamic virtual avatars through lightweight control logic; and conducting multimodal interactive guidance based on the dynamic virtual avatars, verifying the consistency between user input information and virtual avatar actions in real time. This technical solution improves the accuracy of health status reconstruction and the effectiveness of interactive guidance.

This application provides a data processing method and electronic device for geometric components, relating to the field of data processing technology. The method includes: dividing a discrete set of points according to the normal direction corresponding to the geometric component to obtain multiple original layered point sets; sorting the points in each original layered point set according to the multidimensional features of each point to obtain layered ordered point sets corresponding to each original layered point set; constructing a two-dimensional mapping mesh corresponding to each layered ordered point set; mapping each point in each layered ordered point set to the two-dimensional mapping mesh and performing a transformation to obtain the three-dimensional mapping coordinates of each point in each layered ordered point set; and obtaining the mapping result based on the three-dimensional mapping coordinates of each point in each layered ordered point set. This application improves mapping accuracy by adapting complex geometric components to normal layering and combining multidimensional feature sorting, making it applicable to complex component point set processing in multiple fields.

The invention relates to a teaching robot practical evaluation related technology, and provides a teaching robot operation track three-dimensional evaluation method, system and device and a storage medium. The method comprises the following steps: acquiring an image sequence acquired by a fixed external camera in the process of executing a practice task by a teaching robot, and identifying and continuously tracking an end effector, a tool center point or a preset key part to obtain two-dimensional operation track information; mapping the two-dimensional track points into space rays according to internal parameters and external parameters of the camera, and solving a three-dimensional operation track under a world coordinate system by combining a constraint optimization model; and performing action stage division and time sequence alignment according to the key event, the key action node or a preset space region, generating a matching score for each action stage, and outputting a trajectory scoring result. The scheme is suitable for practical task process evaluation in a teaching robot platform.

The application relates to a Web front-end-based interaction method, system and electronic equipment, which calls a terminal camera through a browser, acquires a video stream of a user, and processes the video stream at a browser front end to extract human key points; through a layered action mapping engine, the human key points are mapped into skeleton animation parameters of a virtual image; according to preset attributes of a virtual article and the skeleton animation parameters, the relative poses of the virtual article and the virtual image are calculated; in a browser environment, according to the relative poses, an interaction picture of the virtual article and the virtual image is rendered and displayed. The application solves the problems of high hardware dependency, poor cross-platform compatibility, simple interaction link, and serious delay and distortion in the prior art. Without special hardware, the application reduces the use threshold of users and realizes cross-platform compatibility based on a Web technology stack.