专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明专利

DE3906215A1 未知
公开(公告)号：DE3906215A1
公开(公告)日：1990-08-30
申请号：DE3906215
申请日：1989-02-28
申请人： MASSEN ROBERT PROF DR ING
发明人： MASSEN ROBERT PROF DR ING
IPC分类号： A01G7/00 , B07C5/10 , B07C5/342 , G01B11/00 , G01N21/84 , G06T1/00 , G06T7/00 , G06T7/194 , G06T7/593 , G06T7/90 , H04N7/18 , H04N13/00
摘要： Automatic optical classification of plants in which an image of each plant is produced by a colour video camera. The colour video signals emitted by the colour video camera are digitalized and classified, picture element by picture element, into predetermined colour classes associated with the image of the plant and with the image of the background. A digitized image of the plant characterized by the colour classes is stored in an image memory and evaluated by an image computer in order to classify the plant. The image of the plant is segmented from the image of the background on the basis of the membership of the picture element in the predetermined colour classes, and geometric shape features and colour features are determined from the segmented image of the plant. In addition, the height of the plant is measured, if necessary using two video cameras and a light pattern projector. The colour classifier used for the colour classification is designed so that it can learn the predetermined colour classes through being shown image sections.

52. 发明专利

GB2607385A Generating frames for neural simulation using one or more neural networks 未知
公开(公告)号：GB2607385A
公开(公告)日：2022-12-07
申请号：GB202202537
申请日：2022-02-24
申请人： NVIDIA CORP
发明人： SEUNG WOOK KIM , JONAH PHILION , SANJA FIDLER , ANTONIO TORRALBA BARRIUSO
IPC分类号： G06T11/00 , G06N3/02 , G06T7/194
摘要： Apparatuses, systems, and techniques to use one or more neural networks to generate one or more images 410, 445 based, at least in part, on one or more spatially-independent features within the one or more images 410, 445. In at least one embodiment, the one or more neural networks determine spatially-independent information and spatially-dependent information of the one or more images 410, 445 and process the spatially-independent information and the spatially-dependent information to generate the one or more spatially-independent features and one or more spatially-dependent features within the one or more images 410, 445. The images may be used in a driving simulation system.

53. 发明专利

GB2606807A Image creation for computer vision model training 未知
公开(公告)号：GB2606807A
公开(公告)日：2022-11-23
申请号：GB202201583
申请日：2022-02-08
申请人： AVANADE HOLDINGS LLC
发明人： FERGUS KIDD
IPC分类号： G06T7/194 , G06T11/60
摘要： In some implementations, a device may obtain a set of input images, of an object of interest, that includes images that have a plain background. The device may obtain a set of background images that include images associated with the object of interest. The device may generate, for an input image, a first modified image of the input image that removes a plain background of the input image. The device may generate, for the input image, a second modified image of the input image that is based on the first modified image and a background image. The device may generate, for the input image, a training image that includes an indication of a location of the object of interest depicted in the training image. The device may provide the training image to a training set, that includes a set of training images, for a computer vision model.

54. 发明专利

AU2021204421A1 Data extraction and prediction generation method and system for sporting events 未知
公开(公告)号：AU2021204421A1
公开(公告)日：2022-11-10
申请号：AU2021204421
申请日：2021-06-28
申请人： EDWORKZ PTY LTD
发明人： JOVANOVIC IGGY , DAVARPANAH SAYED HASHAM , GHAZIKHANI ADEL
IPC分类号： A63B24/00 , A63B69/38 , G06T7/194 , G06T7/254 , G06V20/40
摘要： DATA EXTRACTION AND PREDICTION GENERATION METHOD AND SYSTEM FOR SPORTING EVENT A method, processing system, system and computer readable medium is disclosed for extracting analytic data from a video of a sporting event and determining one or more predictions in relation to the sporting event. In one aspect, the method includes steps of: receiving a video of the sporting event; extracting a background image from the video; generating, using the background image, a plurality of moving masks for a plurality of frames of the video; extracting, from the plurality of moving masks, analytic data related to at least a portion of the sporting event; and using the analytic data to generate a likelihood for one or more outcomes for the sporting event, or a portion of the sporting event. Receive a video of the sporting event 310 Extract a background image from the video 320 Generate, using the background image, 330 moving masks for a plurality of frames of the video Extract, from the moving masks, analytic data 340 related to at least a portion of the sporting event Use the analytic data to generate a likelihood 350 for one or more outcomes for the sporting event, or a portion of the sporting event

55. 发明专利

GB2598512B Progressive 3D point cloud segmentation into object and background from tracking sessions 未知
公开(公告)号：GB2598512B
公开(公告)日：2022-10-05
申请号：GB202117072
申请日：2020-05-06
申请人： IBM
发明人： YOCHAY TZUR
IPC分类号： G06T7/194 , G06T19/00
摘要： Embodiments of the present systems and methods may provide techniques that provide automatic, reliable performance of the point cloud cleanup task. Embodiments may provide the capability to progressively learn the object-background segmentation from tracking sessions in augmented reality (AR) applications. The 3D point cloud may be used for tracking in AR applications, by matching points from the cloud to regions in the live video. A 3D point that has many matches is more likely to be part of the object, while a 3D point that rarely has matches is more likely to be part of the background. Advantages of this approach include that no manual work is needed to do the segmentation, and the results may be constantly improved over time, as the object is being tracked in multiple environments.

56. 发明专利

GB2587841B Utilizing a neural network having a two-stream encoder architecture to generate composite digital images 未知
公开(公告)号：GB2587841B
公开(公告)日：2022-09-21
申请号：GB202007986
申请日：2020-05-28
申请人： ADOBE INC
发明人： ZHE LIN , JIANMING ZHANG , HE ZHANG , FEDERICO PERAZZI
IPC分类号： G06T11/60 , G06T7/194
摘要： The present disclosure relates to utilizing a neural network having a two-stream encoder architecture to accurately generate composite digital images that realistically portray a foreground object from one digital image against a scene from another digital image. For example, the disclosed systems can utilize a foreground encoder of the neural network to identify features from a foreground image and further utilize a background encoder to identify features from a background image. The disclosed systems can then utilize a decoder to fuse the features together and generate a composite digital image. The disclosed systems can train the neural network utilizing an easy-to-hard data augmentation scheme implemented via self-teaching. The disclosed systems can further incorporate the neural network within an end-to-end framework for automation of the image composition process.

57. 发明专利

LU102082B1 IMAGE SECURITY USING SEGMENTATION 未知
公开(公告)号：LU102082B1
公开(公告)日：2022-03-29
申请号：LU102082
申请日：2020-09-25
申请人： MICROSOFT TECHNOLOGY LICENSING LLC
发明人： LIN QINGFEN , TURBELL HENRIK VALDEMAR , ZHAO DAVID YUHENG
IPC分类号： H04N21/258 , G06T7/194 , H04N21/2187 , H04N21/4223 , H04N21/4415 , H04N21/442
摘要： Systems and methods may be used for access control. These systems and methods may include using a data processing system to access a video stream, the video stream including an image including a virtual background, segmenting the image into a foreground portion and a background portion to determine whether the foreground portion or the background portion of the image meets a threshold requirement, and outputting an alert in response to determining that the foreground portion or the background portion of the image fails to meet the threshold requirement.

58. 发明专利

BR112021024268A2 Representação e extração de movimento em camadas em vídeos de câmera estática monocular 未知
公开(公告)号：BR112021024268A2
公开(公告)日：2022-01-11
申请号：BR112021024268
申请日：2020-05-18
申请人： EVERSEEN LTD
发明人： CERNAZANU-GLAVAN COSMIN , PESCARU DAN , GUI VASILE
IPC分类号： G06T7/11 , G06T7/143 , G06T7/174 , G06T7/194 , G06T7/215 , G06T7/246 , G06T7/254 , G06T7/277
摘要： representação e extração de movimento em camadas em vídeos de câmera estática monocular. a presente invenção refere-se a um método para segmentar os dados de movimento e inferir a ordem de camadas em um vídeo, que inclui segmentar um primeiro quadro de imagem do vídeo usando estimativa de modo intra-quadro, segmentar quadros adicionais de imagem do vídeo usando correspondência de modo inter-quadro, gerar um mapa de mudança de rótulo para um quadro de imagem segmentada atual gerando um mapa binário do quadro de imagem segmentado atual, e então marcar um ou mais pixels cujos rótulos mudaram com respeito a um quadro de imagem segmentada prévia, detectar um ou mais segmentos de imagem não estática do quadro de imagem segmentada atual com base no mapa de mudança de rótulo, detectar um ou mais segmentos de imagem móvel nos um ou mais segmentos de imagem não estática, estimar o movimento de um ou mais segmentos de imagem móvel, representar cada segmento de movimento como um vetor de sete parâmetros, e agrupar um ou mais vetores de segmento móvel similares com base em um ou mais parâmetros correspondentes.

59. 发明专利

ES2887775T3 Sistema para maximizar dinámicamente el contraste entre el frente y el fondo en imágenes y/o secuencias de imágenes 未知
公开(公告)号：ES2887775T3
公开(公告)日：2021-12-27
申请号：ES17790975
申请日：2017-10-09
申请人： UNIQFEED AG
发明人： DRAGON RALF
IPC分类号： G11B27/031 , G06T7/11 , G06T7/143 , G06T7/194 , G16B20/00 , H04N5/222
摘要： Sistema para la maximización de contraste dinámica entre frente y fondo en imágenes y/o secuencias de imágenes, que comprende al menos un dispositivo (C1, C2) de captación de imágenes, al menos un módulo (A1, A2) de análisis conectado con el dispositivo (C1, C2) de captación de imágenes, que está configurado para determinar una transformación entre un sistema de coordenadas bidimensional del dispositivo (C1, C2) de captación de imágenes y un sistema de coordenadas bidimensional de un dispositivo (L) de pantalla, para realizar una segmentación para determinar en una imagen al menos ocultaciones de objetos en el fondo mediante objetos en el frente, al menos un módulo (K) de control conectado con el módulo (A1, A2) de análisis, al menos un dispositivo (L) de pantalla, el cual está configurado para representar información de visualización, en particular, clips de vídeo, estando el dispositivo (C1, C2) de captación de imágenes configurado captar la información de visualización representada en el dispositivo (L) de pantalla, una base (D) de datos, la cual está conectada con el módulo (K) de control y el dispositivo (L) de pantalla, estando la base (D) de datos configurada para almacenar una secuencia de informaciones de visualización, las cuales deben representarse una tras otra en el dispositivo (L) de pantalla, estando el módulo (A1, A2) de análisis configurado además para determinar datos de calidad acerca de la segmentación y enviarlos al módulo (K) de control, estando el módulo (K) de control configurado para adaptar las informaciones de visualización representadas en el dispositivo (L) de pantalla en base a los datos de calidad, caracterizado por que el módulo (A1, A2) de análisis está configurado para aplicar y minimizar una función de energía en la segmentación, basándose la función de energía en distribuciones de probabilidad de objetos en el frente y objetos en el fondo y/o en distribuciones de probabilidad por pixel, y que el módulo (K) de control está configurado para determinar para informaciones de visualización que permanecen como están de la secuencia, en qué información de visualización se diferencian más una distribución de probabilidad esperada del frente y una distribución de probabilidad esperada del fondo del dispositivo de captación de imágenes.

60. 发明专利

DE112020000906T5 SCHRITTWEISE 3D-PUNKTWOLKENSEGMENTIERUNG IN OBJEKT UND HINTERGRUND AUS ERFASSUNGSSITZUNGEN 未知
公开(公告)号：DE112020000906T5
公开(公告)日：2021-11-11
申请号：DE112020000906
申请日：2020-05-06
申请人： IBM
发明人： TZUR YOCHAY
IPC分类号： G06T7/194 , G06T7/70
摘要： Ausführungsformen der vorliegenden Systeme und Verfahren können Techniken bereitstellen, die ein automatisches und zuverlässiges Durchführen einer Aufgabe zum Bereinigen von Punktwolken bereitstellen. Ausführungsformen können die Fähigkeit bereitstellen, die Objekt-Hintergrund-Segmentierung aus Erfassungssitzungen in Augmented-Reality-Anwendungen (AR-Anwendungen) schrittweise zu erlernen. Die 3D-Punktwolke kann für Erfassen in AR-Anwendungen verwendet werden, indem Punkte aus der Wolke mit Bereichen im Live-Video in Übereinstimmung gebracht werden. Ein 3D-Punkt, der viele Übereinstimmungen hat, gehört mit größerer Wahrscheinlichkeit zum Objekt, während ein 3D-Punkt, der selten Übereinstimmungen hat, mit größerer Wahrscheinlichkeit ein Teil des Hintergrunds ist. Vorteile dieses Ansatzes sind unter anderem, dass keine manuelle Arbeit für die Segmentierung erforderlich ist und die Ergebnisse im Laufe der Zeit ständig verbessert werden können, da das Objekt in verschiedenen Umgebungen erfasst wird.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式