34results about How to "Achieve real-time rendering" patented technology

Provided is a real-time ray cast volume rendering method of three-dimensional earthquake volume data. The compute unified device architecture (CUDA) parallel technology is used for judging whether volume data blocks are visible, then the CUDA parallel technology is used for judging whether current resolution ratios of the volume data blocks need improvement, the two steps start looping from low-resolution, the resolution ratios are improved step by step until resolution ratio requirements are met, or the resolution ratios can not be improved due to insufficient video memory or a highest resolution ratio is achieved, and therefore the resolution ratios respectively used by the volume data blocks in the process of final render can be confirmed. The multithreading technology is used in the process of render processing for carrying out I / O dispatch on the volume data blocks to be dispatched, and meanwhile, the CUDA parallel technology is used for carrying out ray cast volume rendering on dispatched data. Compared with a ray cast volume rendering method in the prior art, the real-time ray cast volume rendering method is applicable to real-time volume rendering of massive volume data and can be used for rendering of data with various resolution ratio grades in a scene, computation speed is fast, and imaging effect is good.

The invention is a method for real-time drawing geometric model of realistic 3D topography, characterized in comprising the following steps of: (1) preprocessing topographical data to its structure accord to the follow-up operation; (2) adopting a topographical planning algorithm of making file-file real-time switching based on traditional LRU algorithm and the method of integrating viewpoint interests to complete the real-time planning for large-scale topographical files; (3) simplifying topographical model based on geometric Mip Map topographical network implementing and generating algorithm;(4) displaying topographical feature based on multi-grain GPU topographical Bump Map rendering algorithm; (5) realizing topographical data conversion based on four-tree ergodic topographical partial-replacing algorithm. The invention solves the problem of the drawing speed of topographical model being low, raising the drawing efficiency and achieving the purpose of real-time drawing dynamical topography.

The invention provides a large-scale scene real-time rendering method based on user behavior analysis, which belongs to the technical field of machine learning. The method first collects user movement behavior data in a virtual reality scene, and constructs a user behavior database in a virtual scene. Then build the basic model of user behavior analysis based on the neural network, and use the user behavior database to train the model. Finally, calculate the user's position and vision range at the next moment according to the personalized model, and output the scene to be rendered according to the current virtual reality scene data. When the amount of models in the scene is large and the model quality is high, this method can pre-load the models within the user's possible vision range at the next moment, realize real-time rendering of large-scale scenes, save rendering resources, avoid rendering delays, and create A truly immersive experience.

The present invention provides a method and device for roaming in a scene with a free view angle. The method includes: inputting the parameter information of the camera array and the position information of the virtual camera into the preset calculation model, and outputting the numbered coordinates of the target camera, where the numbered coordinates are used to represent the target The position of the camera in the camera array; if it is judged that the two numbers in the numbered coordinates are both integers, the target pixel in the image captured by the target camera is used as the virtual pixel in the virtual image captured by the virtual camera, and the target coordinate of the target pixel The same as the virtual coordinates of the virtual pixels, the target pixel value of the target pixel is equal to the virtual pixel value of the virtual pixel, so as to realize the free-view scene roaming. The free viewing angle scene roaming method and device provided by the present invention obtain dense viewpoint information from existing light field information, generate virtual images of virtual cameras, realize free scene roaming within a certain range, and realize real-time rendering.

Provided is a real-time ray cast volume rendering method of three-dimensional earthquake volume data. The compute unified device architecture (CUDA) parallel technology is used for judging whether volume data blocks are visible, then the CUDA parallel technology is used for judging whether current resolution ratios of the volume data blocks need improvement, the two steps start looping from low-resolution, the resolution ratios are improved step by step until resolution ratio requirements are met, or the resolution ratios can not be improved due to insufficient video memory or a highest resolution ratio is achieved, and therefore the resolution ratios respectively used by the volume data blocks in the process of final render can be confirmed. The multithreading technology is used in the process of render processing for carrying out I / O dispatch on the volume data blocks to be dispatched, and meanwhile, the CUDA parallel technology is used for carrying out ray cast volume rendering on dispatched data. Compared with a ray cast volume rendering method in the prior art, the real-time ray cast volume rendering method is applicable to real-time volume rendering of massive volume data and can be used for rendering of data with various resolution ratio grades in a scene, computation speed is fast, and imaging effect is good.

The invention discloses an image registration method and device. The method includes: determining the target plane according to the position of the optical center of the camera in the planar camera array; and orthogonally projecting the optical center of each camera in the planar camera array to the target plane , to get the projection point of the optical center of each camera; according to the projection point of the optical center of the camera in the plane camera array, a rectangular grid is generated on the target plane; according to the rectangular grid, the target grid point of each camera is obtained; The target grid point of each camera is the origin, and the target camera coordinate system of each camera is determined; according to the spatial position relationship between the cameras in the planar camera array and the target camera coordinate system, the images collected by the cameras in the planar camera array are registered . In the embodiment of the present invention, by re-determining the target camera coordinate system of the camera, the complexity of subsequent image registration is greatly reduced.

The present application provides a method and device for improving the rendering efficiency of GIS point data at the browser end, which is characterized in that it includes: storing GIS point data, the GIS point data including identification ID and spatial geographic coordinates; and the corresponding graphic files are encapsulated into a browser-side canvas object; batch rendering is performed on the browser-side canvas object. This application supports commonly used SQL Server, Oracle, Dameng, Jincang and other databases, GeoPackage and UDB database files, as well as GIS point data stored in files such as Excel and CSV, and can realize real-time rendering of a large number of GIS point data on the browser side.

The invention discloses a prediction and real-time rendering system for mandibular angle osteotomy, which comprises a calibration subsystem, a prediction subsystem and a rendering subsystem, and is characterized in that the calibration subsystem is used for obtaining a lower dentition rivet point and a danger area of a mandibular angle osteotomy patient according to a preoperative CT image of themandibular angle osteotomy patient and calibrating a maximum osteotomy range; wherein the prediction subsystem is used for inputting a preoperative CT image, a preoperative photo, a prediction postoperative photo and a maximum osteotomy range of a mandibular angle osteotomy patient, into an osteotomy surface prediction model and predicting an actual patient's cutting line and an osteotomy surface,and the rendering subsystem is used for rendering the surgical field cutting line, the osteotomy surface and a danger area in real time in combination with AR equipment. According to the pre-operation CT, the pre-operation picture and the osteotomy surface prediction model of the actual patient, the operation field of the actual patient is rendered in real time, the operation precision is improved, the operation risk is reduced, the operation time is shortened, and operation complications are reduced.

An adjustable hyperspectral detection chip enhanced by a multi-resonance plasmonic mechanism. The detection chip consists of an array of metal nanonail resonator detection units. Each detection unit (1) comprises: a bottom electrode (2), a semiconductor material layer (3), a spacer layer (4), a nanonail array (5), a control material layer (6), a top electrode (7), a peripheral control signal (8), and a driving circuit (9). The positional relationship from top to bottom is the top electrode (7), the control material layer (6), the nanonail array (5), the spacer layer (4), the semiconductor material layer (3), and the bottom electrode (2). The nanonail array (5) is loaded inside the control material layer (6), and the peripheral control signal (8) and the driving circuit (9) are connected to both sides of the control material layer (6).

The present invention provides a method, device, and electronic device for loading expression resources; wherein, the method includes: obtaining the expression resources of the virtual character in the target animation; the facial expression of the virtual character is composed of at least one expression mixed shape; according to the virtual The degree of influence of the expression blend shape in the character's facial expression on the image of the target animation, determine the expression blend shape that needs to be loaded from the expression resource; load the expression blend shape determined from the expression resource. In this method, according to the degree of influence of the expression blend shape on the target animation, the expression blend shape is selectively loaded, and only the expression blend shape that has a greater impact on the image quality of the target animation can be loaded, and there is no need to load the image that has a greater impact on the target animation. The small expression blend shape not only ensures the playback effect of the animation screen, but also reduces the consumption of memory resources and computing resources of the device, which is conducive to the real-time rendering of high-quality facial expressions.

The invention discloses a method and a device for online preview of a three-dimensional model. The method comprises the steps of locally pre-rendering a picture in multiple view angles for the three-dimensional model; and responding to the online preview of the three-dimensional model. According to the method and the device for the online preview of the three-dimensional model, real data of the model do not need to be loaded locally, a user is prevented from contacting the real data and the data safety is protected; an optimization scheme for rendering of the view angles is provided, thus thenumber of images needing to be drawn is greatly reduced, the data transmission size is decreased and the data transmission time is shortened, the real-time rendering of a three-dimensional scene is realized, and the usage experience of the user is improved; the display performance requirement for mobile terminal equipment is reduced; and the limit of a browser is avoided so that the preview of a large scene becomes possible.