22 results about "Stereoscopic visualization" patented technology

The invention discloses a real-time interaction 3D digital holographic cabin method based on deep learning and sound cloning. The method comprises the following steps that S1, user data are collected and preprocessed; s2, extracting feature vectors of facial expressions and limb actions; s3, generating speech synthesis data by using the improved GE2E network and a preset target speech text; s4, generating a synthesized voice audio based on the voice synthesis data; s5, generating a three-dimensional digital human motion sequence according to the facial expression feature vector and the body motion feature vector; s6, performing timestamp alignment on the three-dimensional digital human action sequence and the synthesized voice audio, and constructing a synchronous output stream; and S7, rendering the synchronous output stream, and performing three-dimensional visual output. According to the method, the improved GE2E network, deep learning and sound cloning methods are fused, three-dimensional virtual human voice action synchronous control is achieved, and the method has the advantages of being high in real-time performance, high in immersion and natural in interaction.

The invention discloses an intraoperative ultrasound and preoperative image non-rigid registration method based on topological high-order MRF (Markov Random Field). The method comprises the following steps: firstly, taking iUS as a target image T, taking a preoperative image as a floating image M, and respectively processing the target image T and the floating image M; then according to the Markov property of the MRF, firstly defining a neighborhood system N and a sub-group C, and then respectively defining context constraints of the processed target image T and the floating image M through the defined neighborhood system N and the defined sub-group C; establishing a non-rigid registration energy function of the target image T and the floating image M based on the topological high-order MRF on the basis of the defined neighborhood system N and the sub-group C; and finally, realizing calculation of a non-rigid registration energy function of the target image T and the floating image M based on the topological high-order MRF based on a multi-resolution additional-variable-free reduced-order energy optimization algorithm. According to the method, the intraoperative dynamic decision-making capability of surgical navigation is improved, precise three-dimensional visual surgical guidance is provided for doctors clinically, and the surgical precision is improved.

Systems and methods employing a surgical robotic surgeon console with a display for use by a surgeon in a remote surgery are provided. The surgical robotic surgeon console includes a console frame and a horizontal member connected to the console frame. the horizontal member extending outwardly from the console frame and positioned above the display. Also provided is a head bar mounted on the horizontal member and the head bar configured to support a user's head relative to the display. Also provided is a system for camera control based at least in part on live gaze location information.

The invention discloses a knowledge three-dimensional topology network modeling and three-dimensional visualization system, which comprises a plurality of supporting plates, a plurality of movable units, a plurality of linking strips, a plurality of sliding guide rods and a plurality of movable bases, and the movable bases are symmetrically distributed in pairs in the vertical direction. Different knowledge display blocks are attached to movable units, a plurality of polyhedral movable units are stopped and attached to a plurality of supporting plates arranged spatially, a plurality of linking strips penetrate through holes of the supporting plates at different positions, and the supporting plates are spatially arranged on sliding guide rods. A single knowledge point can realize rapid position switching and attachment stopping according to the thought of an operator, a plurality of knowledge three-dimensional topology network modeling and three-dimensional visualization systems capable of being manually adjusted are formed through combination by integrating the structure and space corresponding relation of each component, a dynamic three-dimensional association model capable of being built is provided, and the dynamic three-dimensional association model has the advantages of being simple in structure and convenient to use. And the systematic thinking of solving complex problems is cultivated.

Provided is a super-resolution stereoscopic visualization image processing system having a high-speed processing function capable of obtaining a concave-convex image at a high speed with a detailed resolution. The present invention is provided with a base map database (110), a region definition unit (112), a super-resolution rasterizing processing unit (135), a moving average unit (134), a rectangular plane coordinate conversion unit (145), a super-resolution image generation unit (151), an X-direction adjustment unit (152), and a display processing unit (150). An interpolation process is performed, and after a specific number of moving averaging processes are applied to the post-interpolation elevation value (zri) of each super-resolution fine mesh (mbi) of the super-resolution square mesh (Mbi), a planar right-angle super-resolution mesh (Mdi) is generated and a square-adjusted super-resolution mesh (Mei) is generated. Thereafter, a super-resolution image is generated.

A system for guiding an ophthalmic procedure is disclosed. The system includes a housing assembly with a head unit configured to be at least partially directed towards a target site in an eye. An optical coherence tomography (OCT) module and stereoscopic visualization camera are at least partially located in the head unit and configured to obtain a first set and a second set of volumetric data, respectively. A controller is configured to register the first set and second set of volumetric data to create a third set of registered volumetric data. The third set and second set of registered volumetric data are rendered, via a volumetric render module, to a first and second region. The first region and the second region are overlaid to obtain a shared composite view of the target site. The controller is configured to extract structural features and / or enable visualization of the target site.

The application discloses a kind of based on topological high-order MRF intraoperative ultrasound and preoperative image non-rigid registration method, first iUS as target image T, preoperative image is as floating image M and is respectively handled;Then according to the Markov property of MRF, first define neighborhood system N and subgroup C, then define the context constraint of the target image T and floating image M after processing respectively by defined neighborhood system N and subgroup C;Again based on defined neighborhood system N and subgroup C, realize the non-rigid registration energy function of target image T and floating image M based on topological high-order MRF;Finally, based on the calculation of the non-rigid registration energy function of target image T and floating image M based on topological high-order MRF, based on multi-resolution additional variable reduction energy optimization algorithm is realized.The application improves the intraoperative dynamic decision-making ability of surgical navigation, provides accurate stereoscopic visual surgery guidance for physicians in clinic, and improves surgical precision.

The invention relates to a data processing device (300) for a medical viewing apparatus, such as an endoscope, a microscope or any other type of medical imaging device. The data processing device is configured to acquire (110) input image data (115), the input image data representing a scene obtained by the medical observation apparatus, analyze (120) the input image data to determine different categories (125) in the scene, generate (130) a plurality of stereoscopic images (135) from the input image data (115), each of the stereoscopic images representing a different category (125) determined in the scene, and transmit the stereoscopic images (135) to the medical observation apparatus. Based on the determined category, a different disparity is assigned (140) to each of the plurality of stereoscopic images to produce a plurality of processed stereoscopic images (145) and the plurality of processed stereoscopic images (145) are combined (150) to generate a combined stereoscopic image (155). Advantageously, the data processing device (300) enables stereoscopic visualization even if it does not necessarily require input image data (115) from a device compatible with stereoscopic imaging. The invention also relates to a medical observation device (304) having such a data processing device (300). Further, the invention relates to a computer-implemented method (100) as well as a computer-readable medium and a computer program product.

A system for guiding an ophthalmic procedure is disclosed. The system includes a housing assembly with a head unit configured to be at least partially directed towards a target site in an eye. An optical coherence tomography (OCT) module and stereoscopic visualization camera are at least partially located in the head unit and configured to obtain a first set and a second set of volumetric data, respectively. A controller is configured to register the first set and second set of volumetric data to create a third set of registered volumetric data. The third set and second set of registered volumetric data are rendered, via a volumetric render module, to a first and second region. The first region and the second region are overlaid to obtain a shared composite view of the target site. The controller is configured to extract structural features and / or enable visualization of the target site.

The invention relates to the technical field of ship engineering and BIM, and particularly discloses a sand-paving ship construction process quantitative management and BIM data integration system. The system comprises a construction data acquisition module, a three-dimensional thickness modeling and visualization module, a multi-dimensional quantitative evaluation module, a multi-source information fusion analysis module and a process parameter optimization decision module, through integration of high-precision sounding, thickness sensing, equipment operation and geological data, a three-dimensional thickness distribution model is constructed, and a multi-dimensional quantitative index is calculated. And establishing a correlation model of equipment parameters, geological conditions and thickness uniformity, and dynamically generating a process optimization instruction based on an analysis result. According to the invention, three-dimensional visual monitoring, multi-dimensional quantitative evaluation and closed-loop optimization control of the construction quality of the sand-paving ship are realized.

The invention provides an intelligent operation and maintenance management method and system based on an industrial and commercial energy storage cabinet, and the method comprises the steps: deploying a network communication module and a sensor in the energy storage cabinet, monitoring the operation state data in real time, and collecting and summarizing the data through an energy management system, and uploading the data to a cloud platform; historical fault information is analyzed, a fault processing expert knowledge base is established, and state monitoring and fault early warning are carried out on the measuring points prone to faults; when the online operation and maintenance cannot repair the fault, calling real-time meteorological data, and generating an offline operation and maintenance plan in combination with the scheduling information of the operation and maintenance personnel; and constructing a three-dimensional visual model of the energy storage cabinet based on a 3D modeling technology, and marking a fault point for operation and maintenance personnel to contrast and maintain. The operation and maintenance efficiency and reliability of the industrial and commercial energy storage cabinet can be improved, the operation and maintenance cost is reduced, and the fault processing time is shortened.

Ophthalmic Surgery Visualization System (10) with a device (16) for stereoscopic visualization of an object area with an operating site (12) on a patient eye (14), which includes an image acquisition device (33) for acquiring object area image data containing image data for at least one patient eye registration structure (76), with a computing unit (54), and with a biometric data interface (70) for providing patient eye biometric data, which defines a patient eye model (102) with a model registration structure (76') corresponding to the patient eye registration structure (76) and a model coordinate system (77'), characterized by the fact that The computer unit (54) contains a computer program with a program routine for registering the patient eye model (102) to the patient eye (14) using the patient eye registration structure (76) and the model registration structure (76'), in which the object area image data acquired by means of the image acquisition device (33) are linked with optical imaging parameters of the image acquisition device (33) in order to provide a patient eye coordinate system (77) fixed to the patient eye (14) in the form of data, which is referenced to the model coordinate system (77'), wherein a stereoscopic display device (48) is provided for displaying image information in a stereoscopic visual impression, wherein the computer program has a program routine for calculating stereoscopic display information in the patient eye coordinate system (77),wherein the program routine serves to provide display information data which can be displayed at the display device (48) as a spatially extended structure (122) superimposed on the visualized object area, wherein the device (16) for stereoscopic visualization of an object area includes an adjustable imaging optic with an autofocus system (130) for automatically focusing on a structure (76) of the patient's eye or for automatically focusing on a point (140') in the patient's eye (14) corresponding to a structure (76') of the patient's eye model (102) or to a point (140') in the patient's eye (14) corresponding to a point of interest (140) of the patient's eye model (102), which has a focusing state interface (138) for providing z-focusing and / or xy-scaling of the imaging optics as well as xy-positioning of the device (16) for acquiring stereoscopic image data in a plane perpendicular to an optical axis (21) of a main objective (20) of the imaging optics, wherein the computer program receives information from the group z-focusing, xy-scaling of the imaging optics, xy-positioning from the focusing state interface (138) and receives patient eye biometric data from the biometric data interface and takes this into account for calculating the display information data.

The invention is suitable for the technical field of environmental geological information visual modeling, and provides a soil-underground water organic pollutant spatial distribution multi-element stereoscopic technical method, which comprises the following steps: by adopting GBIM data management and a WebGL visual cloud platform as a basic platform, integrating a plurality of functional modules, a three-dimensional geological structure online modeling technology and an organic pollutant transport simulation technology are coupled to realize online modeling of a polluted site, and based on a three-dimensional scene stereoscopic visualization technology, different colors and transparencies are distributed to each body pixel in a body model through a ray casting algorithm; and the multi-element visibility is adjusted by adopting different transfer functions in the rendering process to obtain the spatial distribution effect of the surface and the interior of the organic pollutant. Therefore, the migration track and the multi-element distribution characteristics of the pollutants in the geological body can be displayed more clearly, and scientific, visual and dynamic decision support is provided for site pollution assessment and treatment scheme design.

The invention discloses a music teaching demonstration system, and belongs to the technical field of music education. The system comprises a multi-modal input layer, an intelligent analysis layer, a self-adaptive demonstration layer and a collaborative management platform. The multi-mode input layer acquires playing data in real time through the sensor array, the environment audio acquisition module and the intelligent musical instrument interface. The intelligent analysis layer performs real-time analysis and prediction through an edge calculation unit and a neural feedback prediction engine; the self-adaptive demonstration layer provides multi-modal teaching feedback through an AR rendering engine, a tactile feedback device and a situation generator; and the collaborative management platform realizes cross-device collaboration through the cloud server, the teacher control terminal and the student terminal. According to the invention, stereo visualization, real-time multi-mode feedback and personalized teaching of music elements are realized, and the effect and experience of music teaching are remarkably improved.

The embodiments of the present application disclose a three-dimensional static model reconstruction method and device. The method includes collecting multiple picture sequences of a target object, and performing binary segmentation on each picture sequence to respectively obtain corresponding contour maps; obtaining a three-dimensional visualization shell structure based on the contour maps; obtaining a number of three-dimensional model point cloud data within each contour map in the three-dimensional visualization shell structure based on a preset color consistency calculation criterion; and reconstructing the three-dimensional model point cloud data. The technical solution provided by the embodiments of the present application collects different picture sequences from multiple perspectives for the target object, performs binary segmentation on the picture sequences, then obtains a three-dimensional visualization shell structure, and combines the color consistency information to perform three-dimensional reconstruction on the three-dimensional visualization shell structure. The algorithm of the present application has certain superiority and robustness, can effectively solve the material problem of three-dimensional model printing in additive manufacturing, and provides strong support for obtaining three-dimensional models at low cost.

The present invention discloses a real-time interactive 3D digital holographic cabin method based on deep learning and sound cloning, comprising the following steps: S1, collecting user data and preprocessing it; S2, extracting facial expression and body movement feature vectors; S3, generating speech synthesis data using an improved GE2E network and a preset target speech text; S4, generating synthesized speech audio based on the speech synthesis data; S5, generating a three-dimensional digital human action sequence based on the facial expression feature vectors and the body movement feature vectors; S6, aligning the three-dimensional digital human action sequence and the synthesized speech audio with timestamps to construct a synchronous output stream; S7, rendering the synchronous output stream for stereoscopic visualization output. The present invention integrates the improved GE2E network, deep learning, and sound cloning methods to achieve synchronous control of three-dimensional virtual human voice and action, with the advantages of strong real-time performance, high immersion, and natural interaction.

This invention discloses a multi-view optical three-dimensional visualization gas monitoring system and method, relating to the field of gas leak monitoring. It solves the problems of existing technologies, such as difficulty in simultaneously identifying gas types and visualizing their three-dimensional spatial distribution, and insufficient real-time monitoring accuracy. This invention simultaneously acquires multi-view infrared and visible light images through a multi-view collaborative imaging module, adaptively switching between wide-spectrum and narrow-band filtering states based on spectral anomaly intensity to obtain characteristic images of gas clouds. Based on a multi-view stereo matching algorithm, it constructs a three-dimensional spatial point cloud of the gas cloud and locates the leak source. It extracts spectral features for comparison to identify gas types and invert concentrations. Finally, it fuses gas information with the three-dimensional point cloud and real-world images to achieve three-dimensional visualization output and audio-visual alarm linkage. This invention achieves automated, high-precision, and real-time three-dimensional visualization monitoring of gas leaks.

The invention belongs to the crossing field of ocean science and ocean visualization, and discloses a mesoscale vortex three-dimensional visualization method fusing an ocean three-dimensional thermohaline structure, which comprises the following steps: 1) acquiring ocean reanalysis data, and segmenting the ocean reanalysis data according to an ocean standard layer; 2) carrying out layer-by-layer identification on the segmented data, and carrying out interlayer vortex matching to obtain three-dimensional mesoscale vortex information; 3) based on the three-dimensional mesoscale vortex information, intercepting temperature and salinity information of each layer of the vortex from the reanalysis data set, and cutting temperature and salinity data of each layer according to vortex boundary points; and 4) visualizing the cut temperature-salt data, and connecting the vortex inner diameter extreme points of each layer to obtain the three-dimensional mesoscale vortex. According to the method, the three-dimensional form and the thermohaline structure of the mesoscale ocean vortex can be fused and visualized, and the visual expression of the combination of the mesoscale vortex and other ocean hydrological elements is improved.

To generate a fractal terrain stereoscopic visualization image in which jaggies are absent, the terrain appears to be floating, and fine portions of the terrain are represented in detail. By being constituted of a ground map memory 110, a red stereoscopic image generator 120, an area reading unit 140, a 50 m-DEM thinning converter 160, a miniaturization processor 180, a moving average unit 210, a multiply synthesizer 220, a blurred 5 m-DEM smooth red stereoscopic image generator 250, a fractal stereoscopic image generator 260, a display processor 280, and the like, a fractal terrain stereoscopic visualization image with a stereoscopic effect can be obtained in which jaggies are absent, the terrain appears to be floating, and fine portions of the terrain are represented in detail.

3D display stack integrate Al-generated 3D images with secure optical components for anti-counterfeiting and stereoscopic visualization. Thin-film transistor glass layer with pixel control transistors, TFT polarizer for light filtering, color filter glass layer with RGB subpixels, a CF polarizer for light modulation, a MicroOpticalMaterial layer lenticular lense, microstructured diffraction gratings for parallax-based 3D effects and security, liquid optically clear adhesive bonding TFT and CF layers, optically clear adhesive bonding the MOM to a protective-display layer, enabling stereoscopic viewing. AI3D Generative Pretrained Transformer, trained on proprietary datasets of paired RGB images and ground truth depth maps including legacy 3D formats and synthetic data, generates high-accuracy monocular depth maps from single 2D RGB images using encoder-decoder architecture with optimized loss functions. Security features, micro-text, holograms, UV / IR inks, encoded data, embedded in MOM and 3D images, authenticated offline by Secure Pattern Recognition smartphone application via high-resolution camera capture and computer vision.

The invention relates to the technical field of exhibition display, in particular to a three-dimensional visualization device for intelligent interaction. The device comprises a plurality of display assemblies arranged in a row, and each display assembly comprises a driving unit and a light-emitting unit. Each display assembly is correspondingly provided with a first interaction unit, each first interaction unit comprises a first control module and a sensor, and the sensors and the driving units are in communication connection with the first control modules. Each first interaction unit is correspondingly provided with a projection unit, and the projection unit is connected with the corresponding first control module. During use, the sensor detects that a person enters a preset range, the light-emitting unit ascends and descends by a corresponding preset height, and the projection unit projects a preset image on the ground in the preset range near the display assembly. When a person passes through the multiple display assemblies in sequence, the person can observe displayed data conveniently through combination of height dynamic lifting of the light-emitting body and images projected by the projection unit, and the interaction effect is improved.

To obtain a crustal deformation stereoscopic image processing system that can accurately color-code images based on elevation differences that take into account vertical and horizontal components, even when crustal deformation causes displacement in both vertical and horizontal directions. [Solution] A red stereoscopic map Gbi (image) is generated based on a new DEM of a predetermined area (S12), and a grayscale red stereoscopic map GBgi is generated (S16). On the other hand, the new DEM (post-earthquake) and the pre-earthquake (old) DEM (old DEM) are differentiated (simple difference) to highlight the areas of change (S24). Then, these areas of change are color-coded according to the degree of subsidence and uplift (S26).