34 results about "Path tracing" patented technology

Approaches presented herein provide systems and methods for path tracing using a set of textured spherical surfaces obtained from an importance map for an image. An image representation may be generated using the importance map and evaluated to identify a first set of nodes. The nodes may have associated values, such as luminance values, that may be used to subdivide the nodes into bins to maintain a weighed distribution for the associated values. An array of nodes may be generated for sampling and conversion to a three-dimensional direction that may be applied to one or more lighting effects.

Disclosed approaches provide for interactions of secondary rays of light transport paths in a virtual environment to share lighting contributions when determining lighting conditions for a light transport path. Interactions may be shared based on similarities in characteristics (e.g., hit locations), which may define a region in which interactions may share lighting condition data. The region may correspond to a texel of a texture map and lighting contribution data for interactions may be accumulated to the texel spatially and / or temporally, then used to compute composite lighting contribution data that estimates radiance at an interaction. Approaches are also provided for reprojecting lighting contributions of interactions to pixels to share lighting contribution data from secondary bounces of light transport paths while avoiding potential over blurring.

The invention discloses an automatic driving laser radar simulation method for high-fidelity material modeling, and the method comprises the steps: carrying out the high-fidelity modeling of a complete physical process of a single laser channel, and designing a complete technical chain from laser beam modeling, material reflectivity modeling, light path modeling to post-processing. In laser beam modeling, energy space distribution is accurately depicted by introducing an angle attenuation conical beam model; in material reflectivity modeling, a bidirectional reflection distribution function system covering Lambert body, specular reflection, transmission, retroreflection and general materials is established; in light path modeling, traditional single reflection hypothesis is broken through, time-resolved Monte Carlo path tracking and kernel density estimation are adopted, simulation of multi-path reflection and transmission effects is achieved, traditional model bottlenecks are broken through, complex optical phenomena are simulated, the generation mode has expandability and controllability, rare key scenes are efficiently synthesized, and the method is suitable for large-scale popularization and application. And the method has wide field extension potential and high technical universality.

The invention discloses an unmanned aerial vehicle image-assisted path tracking and obstacle avoidance method and system, and the method comprises the following steps: calling a multi-modal sensor group deployed on an unmanned aerial vehicle to collect environment data in response to a flight task instruction of the unmanned aerial vehicle, and carrying out the cross-modal data fusion of the environment data through an adaptive time synchronization algorithm; and calling a preset generative adversarial network model, performing image generation and illumination compensation on low-light image data in the fused environment data, and generating a target enhanced image close to a visible light environment. According to the invention, through cross-modal learning and self-supervised learning technologies, the perception and data synchronization capability of the unmanned aerial vehicle in a complex environment is improved. According to the invention, the generative adversarial network is adopted for image enhancement, and especially in a low-light environment, the accuracy of obstacle detection is improved; real-time obstacle detection and three-dimensional environment modeling are performed in combination with a deep learning model, and path planning and obstacle avoidance control are optimized through deep reinforcement learning.

The invention discloses a product oil supervision beforehand early warning system and method based on AI and big data fusion, and relates to the technical field of data supervision. Comprising the steps of 1, constructing a real-time monitoring network, deploying a millimeter-wave radar liquid level instrument to track the liquid level change of an oil tank in real time, matching a refueling machine tax control sensor and an AI camera for behavior recognition, monitoring refueling operation in a whole process, and collecting whole-process data, 2, integrating heterogeneous data sources, carrying out cross-department data real-time synchronization through Kafka and DataX, and carrying out real-time monitoring on the whole-process data. The method comprises the following steps: step 1, establishing a supervision data lake, and preprocessing the collected data to form a structured data system covering the whole chain of purchase-sales-storage-transportation, and step 2, fusing a multi-modal AI model to identify non-periodic abnormity of the liquid level and temperature of an oil tank and early warning equipment associated faults according to the preprocessed structured data, the method comprises the following steps: step 1, constructing an oil station three-dimensional model, fusing transaction data, monitoring videos and equipment logs, identifying cheating behaviors, and constructing an oil station-equipment-personnel-transportation knowledge graph for path traceability of quality abnormity and cheating risks, and step 4, constructing a 1: 1 oil station three-dimensional model by adopting laser point cloud scanning, and automatically highlighting a fault area and performing whole-process visual display when early warning is triggered.

The invention discloses an automatic distribution method for patrolling and dispatching personnel based on a knowledge graph. The method comprises the following steps: S1, collecting and preprocessing data information; s2, constructing a heterogeneous knowledge graph, setting node types and generating an attribute matrix and an adjacent matrix; s3, traversing a multi-hop path, calculating an edge weight, and extracting a person post semantic feature vector; s4, introducing avoidance, duty rotation and conflict rules, and performing constraint screening on the semantic feature vectors; s5, scoring the candidate persons based on the post matching degree, the historical frequency and the post rotation period, and outputting a post matching score set; s6, sorting the post matching score set, and generating a sorting list according to the fitness from high to low; and S7, performing path traceability on the primary selection result, extracting associated information, and outputting a recommended result set of the calling personnel. According to the invention, the intelligent, compliant and interpretable distribution of the dispatching process of the patrol personnel is realized, and the personnel selection efficiency and the post matching accuracy are remarkably improved.

This invention relates to the field of unmanned surface vessel (USV) path tracking control technology, specifically disclosing a method, computer equipment, and storage medium for path tracking of USVs based on prioritizing experience replay optimization of near-end policy. The method includes setting up a prioritizing experience replay memory pool, an action space, a state space, and a reward function; constructing a network model based on a near-end policy optimization algorithm, and training the network model using samples from the prioritizing experience replay memory pool; generating current state information based on images acquired by the USV, the current position information of the USV, and the position information of matching points on the reference trajectory; and prioritizing the training of samples with poor action performance based on priority sampling data. This invention employs a prioritizing experience replay memory pool to ensure rapid model convergence, and the network model is constructed based on near-end policy optimization. The use of a state space and reward function tailored to environmental conditions improves the algorithm's adaptability and stability to the environment.

This application provides an image distortion simulation method and electronic device. The method includes: acquiring first image parameters of a source image, second image parameters of a target image, and lens parameters of a camera, wherein the target image is a distorted image of the source image after being imaged by the camera's lens; determining the spatial coordinates of each target pixel in the target image based on the second image parameters; performing reverse ray path tracing on each target pixel based on the spatial coordinates, the first image parameters, and the lens parameters to obtain the mapping relationship between each target pixel and each non-distorted pixel in the source image; assigning color values ​​to each target pixel in the target image based on the mapping relationship, and outputting the assigned target image to improve the accuracy, efficiency, and flexibility of image distortion simulation.

A method for rendering an image of a three-dimensional scene using path tracing. For a current pixel of the image to be rendered using path tracing, an associated pixel is identified, associated with the current pixel, wherein at least one parameter for rendering the associated pixel using path tracing has been previously determined. The at least one parameter for rendering the associated pixel is obtained, and used to determine at least one parameter for rendering the current pixel. The at least one determined parameter is output to control a rendering of the current pixel using path tracing.

This invention relates to a system and method for opinion extraction and evolution path tracking, belonging to the field of film review analysis technology. It includes a data input module, a data processing module, a temporal alignment module, a temporal enhancement module, a deep extraction module, an analysis module, and a visualization module. The data input module acquires long film review text data, temporal metadata, and external event data. The text processing module outputs text feature vectors, temporal decay factors, and external event feature vectors. The temporal alignment module outputs aligned temporal context. The temporal enhancement module performs temporal enhancement attention calculation. The deep extraction module extracts deep temporal features. The analysis module outputs sentiment curves and topic drift detection results. The visualization module generates a dynamic reputation evolution map and a visualization analysis report. This application effectively demonstrates the evolution of opinions in long film reviews.

The present disclosure relates to systems and methods for volume data rendering. The methods may include obtaining an initial image relating to volume data of a subject by performing a path tracing operation on the volume data, obtaining a total sampling operation count corresponding to the initial image, determining, based on the volume data, one or more reference parameters relating to the initial image, and generating, based on historical image data relating to the volume date of the subject, the initial image, the total sampling operation count, and the one or more reference parameters, a target image using an image processing model.

A method for rendering an image of a three-dimensional scene using path tracing. Scene feature data for a pixel of the image to be rendered using path tracing is obtained, indicating visual features of a location of the three-dimensional scene for depiction by the pixel in the image, is obtained. The scene feature data is processed at an artificial neural network, ANN, to determine a budget allocation parameter for rendering the pixel using path tracing. The budget allocation parameter is indicative of an amount of computing resources to be used for rendering the pixel using path tracing. The ANN is trained to determine, from the scene feature data, the budget allocation parameter based on the visual features indicated by the scene feature data. The determined budget allocation parameter is output from the ANN to control a rendering of the pixel using path tracing.

Systems and methods for 3D image scanners for real-time dynamic 3D surface imaging are disclosed. Embodiments of the present systems and methods describe a system and method comprising a first and / or second camera, and a projector, and a processor, the projector projects structured light with a fringe pattern onto a 3D object, the processor is configured to extract a phase map and a texture image from the image, and the processor calculates depth information from the phase map. Embodiments further describe methods and systems to determine wrapped phase from the image using a Hilbert transform, and to generate absolute phase from the wrapped phase using a combination of a quality-guided path following algorithm, a two-wavelength phase unwrapping algorithm, or a Markov Random Field method, and to generate a phase map from the absolute phase to determine depth information of the 3D object. Algorithms using conformal mapping, optimal transport mapping, and Tschupmueller mapping are used to register and track captured 3D geometry surfaces.

The application provides a violent and terrorist video recognition method based on key feature fusion. A key feature database based on violent and terrorist video recognition is built in advance; a to-be-recognized segment video is acquired, and content features and propagation channel data of the to-be-recognized segment video are extracted; wherein the content features include audio features and image features; according to the key feature database, audio key features and image key features of the to-be-recognized segment video are subjected to time-axis synchronous space-time fusion, and synchronous space-time features are generated; according to the propagation channel data, propagation path tracing and synchronous space-time feature authentication of the to-be-recognized video are performed, and a video subject of the to-be-recognized video is determined; according to the video subject, violent and terrorist video identification is performed, and when the video subject is a violent and terrorist video, the to-be-recognized segment video is identified as a violent and terrorist video.

The application discloses an infrared scene real-time light tracing rendering method based on a GPU hardware light tracing pipeline and belongs to the field of infrared simulation of virtual scenes. The application is based on the path tracing principle and takes Unity as a platform to realize real-time rendering of infrared scenes. Based on the infrared physics principle, firstly, real-time rendering of the infrared scene is realized based on the path tracing; on the basis, the GPU hardware light tracing pipeline technology is applied to improve the speed of the infrared path tracing, and the importance sampling and the space-time filtering denoising algorithm are used to guarantee the image quality of the infrared scene, so that the real-time rendering of the infrared scene is realized, the realism and details of the infrared rendering result are greatly improved, and the similarity is higher compared with the actually shot infrared scene image.

The invention discloses an infrared scene rendering method based on rasterization and path tracking hybrid architecture, and belongs to the field of image processing, and the method comprises the steps: building a corresponding heat transport tracking path for each heat transport mechanism in a virtual space; according to all the heat transport tracking paths, determining a comprehensive heat transport environment of a virtual space formed by the combined action of all heat transport mechanisms; rasterization rendering is carried out based on the comprehensive heat transport environment, so that an infrared image of the virtual space under the target virtual camera shooting view angle is obtained; and according to the infrared image, generating an imaging image under the target virtual camera view angle. According to the method, the authenticity of the infrared image is improved, the calculation cost and the visual angle sensitivity are reduced, and the adaptability of the infrared rendering process in a complex scene is enhanced.

Methods are provided for visualizing end-to-end path traversed by a hybrid media traffic flow. Specifically, the methods involve obtaining at least one hybrid flow record that includes information about a source of a traffic flow prior to performing a network address translation for the traffic flow. The traffic flow traverses from the source, along a path in a network, to a destination. The method further involves generating a merged flow data record in which at least two flow data records are merged based on the at least one hybrid flow record. The at least two flow data records include a unicast flow data record and a multicast flow data record based on the traffic flow traversing along the path in the network. The method further includes providing an end-to-end visibility of the path traversed by the traffic flow based on the merged flow data record.

This invention discloses a method and system for optimizing dynamic rendering of virtual scenes based on light field reconstruction, specifically relating to the field of image data processing technology in computational imaging. First, it completes the access of multi-source data, geometric / optical calibration, and affine alignment of timestamps, constructing a unified time base and evidence chain within a sliding window of fixed step size. Then, it establishes a multi-scale spatial parameterized structure within the scene bounding box, using path tracing as the forward model to jointly solve geometry, radiation, and material properties, completing static / dynamic decomposition and fusing color, depth, polarization, and spectral residuals to generate an optical path consistency map. Based on this, it constructs a priority scalar field fused from uncertainty and importance, performing area-level budget arbitration and degradation control, and mapping it to executable control quantities such as sampling density, bounce depth, geometric level, and indirect light order. Finally, it improves cross-frame stability through temporal robust fusion and differential updates, and feeds quality metrics and operational statistics back to the next frame's scheduling to form a closed loop.

The approaches presented here provide systems and methods for path tracing using a set of textured spherical surfaces derived from an importance map of an image. An image representation can be generated and evaluated using the importance map to identify an initial set of nodes. The nodes can have associated values, such as luminance values, which can be used to subdivide the nodes into bins to maintain a weighted distribution for the associated values. An array of nodes can be generated for sampling and conversion into a three-dimensional direction that can be applied to one or more lighting effects.

Methods, systems, and machine learning models for denoising deep images are disclosed. Deep image renders can be generated using Monte Carlo rendering methods such as path tracing. Unfortunately, such renders may have visual noise. A machine learning model according to embodiments can be trained to denoise deep images, and can comprise an embedding sub-model and a denoising sub-model. A computer system can use the embedding sub-model to generate a deep image embedding based on a noisy deep image input using novel local attention mechanisms. The computer system can use the denoising sub-model to denoise the noisy deep image using the deep image embedding. In some embodiments, the computer system can use the denoising sub-model to generate denoised deep images at multiple levels (or “scales”) and combine the denoised deep images to produce an output denoised deep image.

The invention provides an intelligent wind power blade inspection method and system based on image recognition, and relates to the technical field of wind power equipment inspection. Firstly, a wind power blade surface image set including continuous waypoint space position identification and continuous time frame sequence image data acquired by unmanned aerial vehicle inspection is acquired; and performing blade surface vibration response mapping processing on the continuous time frame sequence image data to obtain a blade surface vibration response mapping graph, performing vibration wave propagation path tracking processing on the mapping graph corresponding to the continuous spatial position identifier to generate a propagation path tracking result, and performing structure abnormality positioning processing according to the propagation path tracking result to obtain a blade surface vibration response positioning result. And generating a structure anomaly positioning result containing the vibration wave propagation path anomaly deflection point and the vibration wave energy attenuation anomaly region, and finally generating a wind power blade inspection report containing the spatial position label based on the structure anomaly positioning result. The structure abnormity of the wind power blade can be comprehensively and accurately detected, and the inspection efficiency and accuracy are improved.

Methods, systems, and machine learning models for denoising deep images are disclosed. Deep image renders can be generated using Monte Carlo rendering methods such as path tracing. Unfortunately, such renders may have visual noise. A machine learning model according to embodiments can be trained to denoise deep images, and can comprise an embedding sub-model and a denoising sub-model. A computer system can use the embedding sub-model to generate a deep image embedding based on a noisy deep image input using novel local attention mechanisms. The computer system can use the denoising sub-model to denoise the noisy deep image using the deep image embedding. In some embodiments, the computer system can use the denoising sub-model to generate denoised deep images at multiple levels (or "scales") and combine the denoised deep images to produce an output denoised deep image.

The invention discloses a handicraft design method based on water flow numerical model simulation, which relates to the technical field of process design, and comprises the following steps: taking structured flow field gradient data as geometric boundary evolution driving information, and carrying out continuous iteration updating on a parameterized initial geometric matrix by using an implicit topology evolution algorithm to obtain a parameterized initial geometric matrix; enabling the geometric boundary to generate directional evolution under the driving of the flow field gradient until a preset stable condition is reached, forming a stable three-dimensional topological structure model, and calling a flow direction vector field in the structured flow field gradient data to perform surface path tracking and parameter mapping on the basis of the stable three-dimensional topological structure model, projecting the flow direction path to the surface of the model and generating a continuous texture structure to obtain a textured three-dimensional structure model; according to the method, the implicit topology evolution algorithm is utilized to perform continuous iteration updating on the parameterized initial geometric matrix, so that the geometric boundary can automatically generate directional evolution under the driving of the flow field gradient, and the structural form can be gradually optimized along with the change of the flow characteristic.

Ray tracing and path tracing systems employ vertex compression, including per-component shifts, unsigned arithmetic deltas, and base vertex shortening. Topology encoding includes enhanced explicit vertex indexing and implicit, triangle-strip-based indexing, including left, right, start, reverse, and degenerate tokens, as well as rotation. Significant improvements in the rate of lossless compression are achieved, particularly for highly quantized vertex data.

The invention belongs to the technical field of measurement, and provides an imaging model and a calibration method of a spherical mirror image visual device. An imaging model is established based on light path tracing and comprises a forward imaging model and a backward imaging model, and an accurate bidirectional mapping relation between a three-dimensional space point and an image projection point is constructed by tracking incident light and reflected light reflected by a spherical mirror. The calibration method comprises a camera calibration stage, an initial calibration stage and an optimization calibration stage. In the camera calibration stage, a calibration target is used to calibrate an equivalent focal length and a principal point coordinate of an industrial camera; in the initial calibration stage, a spherical mirror outer circle contour projection image is used to calibrate coordinates of a spherical mirror curvature center; in the optimization calibration stage, a calibration target is used, and the coordinates of the curvature center of the spherical mirror are calibrated through a nonlinear optimization algorithm based on the imaging model and the result of the initial calibration stage. According to the invention, accurate bidirectional mapping between three-dimensional space points reflected by the spherical mirror and image projection points is realized.

A method for rendering an image of a three-dimensional scene using path tracing. For a pixel of the image to be rendered using path tracing, a budget allocation parameter for rendering the pixel using path tracing is determined. The budget allocation parameter is indicative of an amount of computing resources to be used for rendering the pixel using path tracing. The budget allocation parameter is determined to optimise the entropy of the image of the three-dimensional scene generated using the rendered pixel. The determined budget allocation parameter is output from the ANN to control a rendering of the pixel using path tracing.

The invention discloses a rice transplanter operation row accurate identification and self-adaptive tracking control method and system, and belongs to the technical field of agricultural machinery intellectualization. The method comprises the following steps: accurately identifying seedling rows through an improved YOLOv8n visual algorithm, wherein a P2 small target detection layer is added, a C2f module is reconstructed by adopting RFAConv, and an SPPF module is replaced by utilizing a Focal Modulation module; performing seedling row positioning point clustering and navigation line extraction based on a detection result, and converting image coordinates into world coordinates through a BP neural network model; and performing path tracking by adopting fuzzy prediction function control based on feedback linearization, performing accurate linearization on a non-linear motion model of the rice transplanter through feedback linearization, and dynamically adjusting a weighting coefficient by utilizing prediction function control taking Morlet wavelet as a primary function and a fuzzy controller. According to the method, the problems of difficulty in seedling row identification and poor path tracking precision in a complex paddy field environment are effectively solved, and the automation level and the operation quality of the rice transplanter are remarkably improved.

Embodiments of the present application disclose an electronic evidence intelligent examination and tracing method, device and related equipment. The method preprocesses the electronic evidence package to be examined, analyzes it into an image stream, an audio stream, a text stream and metadata; performs single-mode authentication analysis in parallel, and performs deep authentication on images, audio, text and video respectively; performs conflict detection on the execution logic of the metadata; cross-verification is performed on the content of the multi-modal data; based on the authentication results, conflict detection results and cross-verification results of the various modal data, a speculative path graph is constructed and generated path tracing is performed to obtain the tracing results. The present application can break through the limitation of a single mode, dynamically detect multiple modal information, trace the source of the evidence, and significantly improve the accuracy and efficiency of the electronic evidence authenticity examination.

The application discloses a kind of unknown environment under mobile robot autonomous exploration method and system, this method first real-time scanning environment obtains the positioning and point information of robot, updates 3D grid map of three-dimensional space representation;Then, boundary detection and candidate viewpoint sampling are carried out on the map, and the weight of viewpoint is calculated to screen expansion node;HybridA* algorithm is used for path planning, and the path is maintained into topological map;Pure path tracking method based on ray distance is used for path tracking, and PID algorithm is used to adjust speed, to improve the accuracy and real-time performance of path tracking.The application improves the completeness of the selection of expansion viewpoint, reduces the repeated exploration of the same area, and improves the path tracking accuracy, effectively solves the technical problems of robot autonomous exploration in large-scale unknown environment.