4137results about "Image generation" patented technology

The invention relates to the technical field of medical rehabilitation, and discloses a lower limb weight-bearing gait rehabilitation training system which comprises a data acquisition module, a data processing and analysis module, a patient individualized modeling module, an intelligent decision and control module, a rehabilitation execution module and a man-machine interaction and medical information interface module which are in communication connection through a network. The data acquisition module is used for acquiring multi-modal data of a patient in real time, and the multi-modal data comprises static sign data, dynamic physiological parameters, kinematics and dynamics parameters and non-motion physiological and psychological state data; and the data processing and analysis module is used for carrying out preprocessing, feature extraction and deep analysis on the original data, and outputting a structured patient individualized feature vector and an evaluation result. According to the invention, a patient three-dimensional skeletal muscle digital twinborn model is constructed through the patient individualized modeling module, and in combination with a continuous learning intelligent model library, body sign differences of different patients can be accurately adapted.

A system for computer assisted navigation during surgery includes a computer platform that operates to register a target surgical area of a patient. In certain cases, a process includes: obtaining a pre-op CT image of a pelvic region of a patient and intra-operatively obtaining a point cloud data about the pelvic region with a navigated instrument, generating a 3D bone model which excludes non-targeted area such as a femur, and then merging the 3D bone model to the point cloud to register the target surgical area.

A programmatically generated AI avatar includes a customizable personality module, acting as the embodied interface for a powerful AI “mind” that delivers personalized coaching to improve user health, well-being, and longevity. The system uses machine learning, large language models, and biometric modeling to synthesize real-time, multi-modal health data—including sleep, nutrition, glucose, mood, and activity—and generate forward-prescribed KHAs. Unlike human coaches, it continuously adapts based on context and behavior, targeting the root cause: metabolic dysfunction—namely by restoring healthy, sustainable body composition through the preservation or building of lean muscle mass and reduction of excess fat. KHAs can also be shared with friends or programmatically generated AI avatars, allowing for coordinated action, emotional support, and accountability through social connection—further reinforcing positive behavior and adherence. The system's reinforcement learning engine incorporates both individual response data and anonymized population-level insights to optimize recommendations over time, learning which interventions are most effective for users with similar physiological and behavioral profiles. First validated with Olympic athletes—resulting in measurable improvements and medal-winning outcomes—this system offers a scalable, emotionally intelligent coaching engine that exceeds human capability, designed for the ultimate purpose of supporting sustainable health, resilience, and human thriving.

The invention provides an unmanned aerial vehicle pose visual angle optimization method and system for fracture refined shooting. The method comprises the steps of performing fracture detection and boundary extraction on a coarse inspection image; recovering a camera track and sparse point cloud based on multi-view three-dimensional reconstruction, carrying out back projection and estimating a normal vector of a crack surface; constructing a shooting spherical shell with limited inner and outer radiuses by taking the crack point as a center, and generating a view cone and spherical shell intersection region allowed to be shot as a candidate set in combination with a normal vector; sampling in the candidate area to generate a plurality of candidate shooting points, and synchronously resolving the flight and holder integrated pose of the corresponding unmanned aerial vehicle; constructing a multi-target cost function including path length, attitude, pan-tilt angle and shooting error, and generating an optimal shooting point sequence and an inspection path through an optimization algorithm; the system realizes fine, efficient and automatic shooting of cracks in a complex structure environment through cooperation of multiple modules, and effectively improves the imaging quality and the detection precision.

A method and system provide the ability to segment a first point cloud. The first point cloud is rendered into multiple two-dimensional (2D) images. The images are segmented to generate a semantic segmentation mask. The images are then backprojected into a 3D classified point cloud. The classified point cloud is segmented into geometric segments and voting is performed for each segment to determine the majority classification and reassign minority classifications. A final point cloud is then exported as a segmented classified point cloud.

The invention provides an airspace simulation deduction method and system based on digital twinning, and belongs to the technical field of airspace simulation, and the method comprises the steps: integrating geographic bottom plate data and airspace element data, carrying out the meshing based on an airspace grid engine, and constructing an airspace digital twinning model; receiving at least one flight plan, and mapping the flight plan to the airspace digital twin model to generate a corresponding simulation aircraft entity; based on a preset simulation engine, driving the simulation aircraft entity to perform flight simulation in the airspace digital twin model, and calculating a simulation flight situation; in the simulation process, airspace conflict detection and early warning are carried out based on the simulation flight situation and the airspace digital twin model; and outputting a simulation deduction result, performing three-dimensional visual output on the simulation flight situation data and the early warning information, and generating an analysis report including conflict statistics and airspace evaluation. The airspace resource allocation is optimized, the flight safety is improved, and intelligent and efficient management is realized.

A method and system for constructing a three-dimensional (3D) aerial survey of a city street scene include obtaining a plurality of video frames from a calibrated multi-camera setup covering a 360-degree view mounted on a moving vehicle. The plurality of video frames is split into a plurality of 3D parts containing a subset of the plurality of video frames and preprocessing the subset of the plurality of video frames of each part of the plurality of parts to obtain a calculated information. Further, constructing, by the processing circuitry, a 3D representation of each part of the plurality of parts based on the calculated information to obtain a plurality of local 3D reconstructed scene intervals. The method includes stitching and filtering, by the processing circuitry, the plurality of local 3D reconstructed scene intervals to construct the 3D city street scene.

A video generation framework that is controllable, unsupervised and based on events (CUBE) includes an event camera, which captures changes in light intensity at each pixel of a scene asynchronously and generates event camera data. A text-to-image diffusion model that is conditioned on textual descriptions integrates the event camera data to control video synthesis. Further, an edge extraction module translates event data into a format usable by the text-to-image diffusion model, whereby the diffusion model synthesizes detailed and contextually accurate videos based on textual prompts. Further, an improved system (CUBE Plus) includes a content frame identification module which selectively identifies and uses only the most information-rich event segments of the event camera data to drive cross-frame attention, and an event driven attention mechanism that allows the framework to focus on event-dense moments.

Techniques are described for automatically generating visual model representations of buildings based at least in part on captured external imagery of the buildings, and using the generated building visual models to generate and present new images and optionally in additional manners, such as to improve navigation of a building and / or its surroundings. The described techniques may include acquiring building data from a plurality of exterior acquisition locations at multiple heights and view angles of an exterior of a building (e.g., using a flying drone and / or other flying device that captures the data), generating visual model representation(s) of the building (e.g., a 3D Gaussian Splat model), and using the generated visual model representation(s) to generate and present a new image with a view of the building exterior from a particular pose along with associated user-manipulatable controls.

A set of images of a scene re received. Each image includes temporal data and spatial data relating to the scene. Based on the spatial data of each image, three-dimensional (3D) Gaussian splatting data is generated. The temporal data of each image and the 3D Gaussian splatting data are inputted to a neural network to generate spatial-temporal 3D Gaussian embeddings. Offset data based on the spatial-temporal 3D Gaussian embeddings is generated. The video of the scene is rendered based on the 3D Gaussian splatting data and the offset data, allowing for improved rendering of video of the scene.

The invention provides a power equipment three-dimensional dynamic updating method and system for virtual reality training, and the method comprises the steps: collecting the working state data of power equipment in real time through a sensor; processing the working state data based on a preset function model, and calculating the physical state change of the power equipment; according to the physical state change, adjusting physical attributes of the three-dimensional model of the power equipment in real time so as to simulate physical effects of thermal expansion, stress concentration or structural deformation; loading the adjusted three-dimensional model into a virtual reality environment, and performing real-time display through a virtual reality rendering platform; and after the user executes the interactive operation through the virtual reality equipment, the working state of the power equipment is updated, and the influence of the operation on the three-dimensional model of the equipment is dynamically fed back, so that a virtual-real linkage closed-loop training process is formed. The virtual model can change in real time along with the collected data, and the technical verisimilitude and operability of virtual training are remarkably improved.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output image using a text-to-image model and conditioned on both the input text and image and text pairs selected from a multi-modal knowledge base. In one aspect, a method includes, at each of multiple time steps: generating a first feature map for the time step; selecting one or more neighbor image and text pairs based on their similarities to the input text; for each of the one or more neighbor images and text pairs, generating a second feature map for the neighbor image and text pair; applying an attention mechanism over the one or more second feature maps to generate an attended feature map; and generating an updated intermediate representation of the output image for the time step.

The invention discloses a geological disaster emergency response path planning method based on reinforcement learning. The geological disaster emergency response path planning method comprises the following steps: S1, fusing and processing multi-source data; s2, constructing a four-dimensional space-time model based on the fused data; s3, multi-agent collaborative decision making is carried out, and a mixed agent system based on a Safe-PPO improved algorithm is deployed; a multi-objective optimization algorithm is adopted to balance the rescue efficiency and safety; deciding and outputting a leading path and a plurality of alternative paths; s4, performing conflict detection on the action of the path planning agent; s5, carrying out preferential selection on the paths subjected to conflict detection and global optimization; s6, issuing the path to a terminal for rescuing the action personnel; s7, dynamically monitoring the environment; s8, judging whether a path needs to be re-planned or not according to environment dynamic monitoring data; and S9, completing the task. According to the method, through reinforcement learning and multi-agent collaborative architecture, the risk penalty term is embedded in the reward function through the improved Safe-PPO algorithm, and the accuracy of path planning is improved.

The invention discloses a mechanical production practice virtual simulation method and system constructed based on a three-dimensional scene CAD model. Firstly, a target workshop is subjected to systematic aerial photography through an unmanned aerial vehicle to collect multi-view image data; a high-fidelity and interactive three-dimensional virtual scene model is automatically constructed by adopting a three-dimensional reconstruction technology fusing generative artificial intelligence and Gaussian splashing; and finally constructing an immersive virtual simulation platform. The platform not only can provide VR roaming and equipment simulation operation functions from a first person perspective, but also can collect user multi-dimensional operation behavior data in real time, and generates a visual intelligent comprehensive evaluation report through a preset multi-dimensional skill map and a quantitative scoring algorithm. According to the invention, high-efficiency and high-fidelity digital reproduction of a mechanical production practice scene is realized, a novel practical teaching mode which is safe, quantifiable and accurate in evaluation is created, and systematic pain points such as single scene, high cost, safety risk and subjective evaluation in traditional practice are effectively solved.

A material for a display mesh is determined by setting a plurality of material IDs for each of a plurality of polygons included in the display mesh. A material for a determination mesh is determined by setting a single material ID for each of a plurality of polygons included in the determination mesh. Voxel data further includes a forced change flag for each voxel. A material for the display mesh and a material for the determination mesh are determined, assuming that a voxel whose forced change flag is on has only a material corresponding to the forced change flag irrespective of the material ID set for the voxel.

Apparatuses, systems, and techniques to predict forces associated with an object's surface. In at least one embodiment, forces associated with an object's surface are predicted using factorized implicit global convolution and one or more neural networks.

Described is a system for generation location visualization on a map interface by identifying a current location of a user that is initiating an interaction function of an interaction client; identifying a map corresponding to the current location of the user; identifying one or more map tiles associated with the map; receiving historical location data of the user that is associated with the current location of the user; converting the historical location data into an overall polygon that is comprised of a plurality of polygons based on the identified one or more map tiles; and displaying the map with the plurality of polygons on a user interface.

The invention discloses an abdominal CT image multi-view fusion method based on HU physical characteristic guidance, and relates to the technical field of medical image processing. The method comprises the following steps: firstly, constructing a three-dimensional HU volume, generating a semantic mask covering the whole HU range as physical prior, and obtaining two-dimensional slice sequences in the axial direction, the coronal direction and the sagittal direction through three-view projection; on the basis, multi-scale deformable alignment, HU-perceived deformation field fine correction, cross-view attention fusion and regional differentiation decision fusion are sequentially carried out, HU-perceived high-frequency residual enhancement and local contrast self-adaptive processing are carried out on two-dimensional slices, and finally a denser CT slice sequence is generated. According to the method, HU physical partition constraint is introduced in the multi-view alignment and fusion process, abnormal deformation and high-density structure distortion of a gas region are effectively inhibited, and geometric consistency and visual stability of an abdominal CT image in the interpolation reconstruction process are improved.

The invention discloses a scene generation method and device based on multi-source GIS data fusion, and relates to the technical field of digital twinning and programmed generation. The method comprises the following steps: acquiring GIS data, preprocessing the GIS data, and storing the preprocessed GIS data in a geographic information resource library; a structured resource library is constructed, semantic parameters are added to the three-dimensional model in the structured resource library through the configuration file, and three-dimensional model resources with structured semantics are constructed; on the basis of the three-dimensional model resources with structured semantics and GIS data in a geographic information resource library, building and road generation and terrain processing are carried out in a programmed modeling engine through a configuration file, and scene data are generated; and importing the generated scene data into a real-time rendering engine, carrying out dynamic environment interaction and biocenosis simulation, and generating a city scene. The problems that in the prior art, an urban three-dimensional modeling method is low in efficiency and insufficient in environment interaction reality sense are solved.

The invention discloses a high-precision and automatic dental crown generation method. The method comprises seven steps of oral cavity three-dimensional data acquisition, full-mouth tooth semantic segmentation, upper and lower jaw point cloud precise registration, abutment and associated gingiva precise segmentation and trimming, abutment surrounding environment multi-tissue segmentation, neck-edge line secondary precise detection, and dental crown generation and multi-dimensional refinement. The method comprises the steps of oral cavity three-dimensional data acquisition, full-mouth tooth semantic segmentation, upper and lower jaw point cloud precise registration, abutment and associated gingiva precise segmentation and trimming, abutment surrounding environment multi-tissue segmentation, neck-edge line secondary precise detection and dental crown generation. Through combination of deep learning and a point cloud algorithm, high-precision segmentation of teeth and gingiva and precise reduction of an occlusion relationship are realized, an oral anatomical feature library and a clinical repair standard are fused, and through multi-dimensional fine adjustment optimization, a dental crown 3D model which is high in fitting degree, harmonious in occlusion and capable of meeting clinical requirements is generated. The method is full-process automatic, greatly improves efficiency, reduces operation threshold, reduces material waste and diagnosis and treatment cost, remarkably improves repair success rate, and is suitable for various oral repair scenes.

A method for encoding three-dimensional (3D) content represented by a dynamic mesh includes: converting geometry information from a list of vertices of the dynamic mesh to a sequence of geometry component images; encoding the sequence of geometry component images of the dynamic mesh using a video encoder to generate a geometry component bitstream; decoding the geometry component bitstream to generate reconstructed geometry component images; determining a face to be removed from connectivity component images of the dynamic mesh, the face containing a vertex in the reconstructed geometry component images; updating the connectivity component images of the dynamic mesh by removing the face from the connectivity component images; encoding the updated connectivity component images to generate a connectivity component bitstream; updating, based on updating the connectivity component images, attribute component images and mapping component images of the dynamic mesh; encoding the updated attribute component images to generate an attribute component bitstream.

The invention discloses a power transmission tower corrosion damage assessment operation and maintenance method and system combining digital twinning and AI, and relates to the technical field of power transmission and transformation project operation and maintenance safety assessment. Based on a parameterized digital twinning base model, a mechanical response and an environmental corrosion rate are obtained by combining environmental parameters, and then a training data set is constructed; training a multi-task machine learning model based on the training data set to obtain a trained multi-task machine learning model; performing feature extraction on the image data to obtain visual features, performing feature extraction on the point cloud data to obtain geometric features, and fusing the visual features and the geometric features by adopting a double-flow feature fusion network based on an attention mechanism to obtain fused features; and inputting the fusion features into a multi-task machine learning model for prediction to obtain a predicted mechanical response and an environment corrosion rate, and then evaluating the corrosion state and predicting the service life so as to realize rapid, accurate and intelligent evaluation and operation and maintenance decision support of the corrosion state of the power transmission tower.

A mesh decoder reconstructs geometry information of a dynamic mesh from a coded mesh bitstream of the dynamic mesh. The reconstructed geometry information include data specifying vertices of the dynamic mesh. The decoder also reconstructs connectivity information of the dynamic mesh which includes data specifying faces of the dynamic mesh. The decoder refines the reconstructed connectivity information based on the reconstructed geometry information to generate refined connectivity information. The decoder further reconstructs an attribute image of the dynamic mesh from the coded mesh bitstream which includes image content to be applied to faces of the dynamic mesh. The decoder refines the reconstructed attribute image based on the reconstructed geometry information to generate refined attribute image. Based on the reconstructed geometry information, the refined connectivity information, and the refined attribute image, the decoder reconstructs the dynamic mesh which can be rendered for display.

Systems and methods are disclosed related to a 3D grounded video foundation model. A video generation method and system provide 3D conditioning information to a video diffusion model to improve generated video quality (object and temporal consistency) that is grounded in three dimensions (3D). The video generation method and system also enable precise camera control, cinematic effects, and scene editing. Video output corresponding to a set of camera specifications is generated for a scene from input image(s) including one or more images of a static scene or a sequence of images (video) for a dynamic scene. The input image(s) are used to calculate a 3D cache representing the scene. The 3D cache is rendered according to the set of camera specifications to produce a frame sequence and a mask sequence that identifies missing pixels in each frame. The frame sequence is encoded and masked to generate the output video.

A computer-implemented method for decoding a coded mesh bitstream of a dynamic mesh representing three-dimensional content includes that: geometry information of the dynamic mesh is reconstructed from a geometry component bitstream in the coded mesh bitstream, the reconstructed geometry information includes data specifying vertices of the dynamic mesh; connectivity information of the dynamic mesh is reconstructed from a connectivity component bitstream in the coded mesh bitstream, the reconstructed connectivity information includes data specifying faces of the dynamic mesh; the reconstructed connectivity information is refined based on the reconstructed geometry information to generate refined connectivity information by at least dividing a face specified by the reconstructed connectivity information into two faces based on a vertex specified in the reconstructed geometry information; the dynamic mesh is reconstructed based on the reconstructed geometry information and the refined connectivity information; and the reconstructed dynamic mesh is caused to be rendered for display.

The invention discloses a geometric constraint fitting point cloud filtering method for sea surface three-dimensional reconstruction, and belongs to the technical field of computer vision and three-dimensional reconstruction. The objective of the invention is to solve the problem of insufficient subsequent three-dimensional reconstruction precision caused by interference of reflection noise, mismatching points and the like in sea surface point cloud. The method specifically comprises the following seven steps: firstly, acquiring sea surface original point cloud through three-dimensional data acquisition equipment; a filtering technology is adopted to obtain a to-be-fitted point cloud; fitting a quadric surface through an improved RANSAC (Random Sample Consensus) algorithm to solve an initial parameter; constructing a comprehensive error function, and optimizing the model through gradient descent; effective inner points are screened through quadratic term coefficient constraint and a distance threshold value; and finally, iterating until a termination condition is met, and outputting an optimal effective point cloud. The method is high in noise rejection rate, the point cloud fits the sea surface form, and high-quality data support can be provided for sea surface fitting, sea wave simulation and unmanned ship control.

A method includes obtaining video data of a dental patient. The method further includes obtaining an indication of selection criteria in association with the video data. The selection criteria includes conditions related to a target dental treatment of the dental patient. The method further includes performing an analysis procedure on the video data. Performing the analysis procedure includes determining a first score for each frame of the video data based on the selection criteria. Performing the analysis procedure further includes determining that a frame satisfies a first threshold condition based on the first score. The method further includes providing the first frame as output of the analysis procedure.

Technology is described for handling truncated data in iterative reconstruction. A method comprises iterating on a volume of an object including a non-truncated part based on image data and at least one truncated part representing deficiently imaged data. The volume is represented by voxels. The iterating includes regularizing the non-truncated part of the volume using a first regularizer, and regularizing the truncated part of the volume using a second regularizer different from the first regularizer.

The invention discloses an intelligent detection method and system for a flat curbstone disease based on linear continuity constraint. The method comprises the following steps: acquiring a road video stream and a vehicle dynamic parameter; extracting the area of the flat curbstone and the surrounding road surface from the road video stream to obtain two-dimensional image information; converting the two-dimensional image information into a visual mimicry geometric model which has a space proportion and represents a three-dimensional geometric contour; constructing a three-dimensional reference model conforming to the actual road alignment based on the visual mimicry geometric model; comparing the spatial difference between the visual mimicry geometric model and the three-dimensional reference model, and calculating and identifying potential diseases to obtain a residual error type, a residual error value and geometric parameters; classifying diseases, evaluating risk levels, and generating maintenance suggestions. By implementing the method provided by the invention, intelligent identification, classification and degree evaluation of the flat curbstone diseases can be realized, the accuracy and automation level of flat curbstone disease detection are improved, and road maintenance is promoted to develop towards a more intelligent and refined direction.

Embodiments of the present disclosure relates to at least one processor including one or more circuits to implement a generative geometry network and an appearance network. The generative geometry network includes a first diffusion model conditioned on at least one input image, the first diffusion model configured to generate a first voxel grid having a first resolution, and a second diffusion model conditioned on the first voxel grid. The second diffusion model configured to generate a second voxel grid having a second resolution. The second resolution is greater than the first resolution, the first voxel grid and the second voxel grid represent a three dimensional (3D) scene. The appearance network predicts one or more Gaussian attributes within one or more voxels of the second voxel grid, determines a representation of a portion of the 3D scene that corresponds to a sky using the at least one input image, and composes a novel view of the 3D scene based at least in part of the Gaussian attributes and the representation of a portion of the 3D scene that corresponds to a sky.