专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

11. 发明专利

GB2591695A Hash-based media search 未知
公开(公告)号：GB2591695A
公开(公告)日：2021-08-04
申请号：GB202105482
申请日：2014-06-24
申请人： GRASS VALLEY LTD
发明人： JONATHAN DIGGINS , GERARD PHILLIPS
IPC分类号： G06K9/00 , G06F16/583 , G06F16/783 , H04N21/235
摘要： A system for identifying video content, comprising: a query sequence processor configured to define a plurality of image sequences by a traveling window that encompasses a plurality of temporally separated video frames, the windowed frames being processed to form a spatial hash value 15, relating to differences between pixels within each image of the windowed sequence, and a temporal hash value 14, relating to differences between pixels of temporally separated frames of the windowed sequence; a database that stores the hash values in various ordered look-up tables 20, 30; and a search processor configured to search the ordered data 34, 24 for a sought hash value. The spatial and temporal hash values may comprise an entropy function and a frequency function. The relational database may comprise: an index ordered 12 first look-up table 10 of hash values and time codes 13; a second look-up table 20 ordered by one of the hash values; and a third look-up table that comprises single values for the ordering hash of the second table and a range of rows where that value of has is present in the second table.

12. 发明专利

AU2020289873B1 User interfaces for selecting media items 未知
公开(公告)号：AU2020289873B1
公开(公告)日：2021-07-08
申请号：AU2020289873
申请日：2020-12-18
申请人： CANVA PTY LTD
发明人： SHEPPARD ELIJAH ALEXANDER , SU YU , KNITTEL PAUL KARL , CHASHCHIN DENIS
IPC分类号： G06F3/0481 , G06F3/0482 , G06F16/44 , G06F16/483 , G06F16/54 , G06F16/583 , H04N21/431
摘要： Abstract Described herein is a computer implemented method. The method comprises displaying a first set of media item representations in a media item order in an initial media selection interface. While the initial media selection interface is displayed, the method further comprises detecting a display complete media set at reference input and, in response: determining a reference media item representation and displaying a complete media selection interface in a first initial display state which is based on the reference media item representation.

13. 发明专利

AU2020202658A1 AUTOMATICALLY DETECTING USER-REQUESTED OBJECTS IN IMAGES 未知
公开(公告)号：AU2020202658A1
公开(公告)日：2021-02-11
申请号：AU2020202658
申请日：2020-04-21
申请人： ADOBE INC
发明人： LIN ZHE , COHEN SCOTT , LING MINGYANG
IPC分类号： G06K9/46 , G06F16/583 , G06K9/62 , G06T7/90
摘要： The present disclosure relates to an object selection system that accurately detects and optionally automatically selects user-requested objects (e.g., query objects) in digital images. For example, the object selection system builds and utilizes an object selection pipeline to determine which object detection neural network to utilize to detect a query object based on analyzing the object class of a query object. In particular, the object selection system can identify both known object classes as well as objects corresponding to unknown object classes. Identify A Query Object To Be Detected Within An Image 202 -1- "Sign" Determine If The Query Object Corresponds To A Known Object Class Object Classes "Sign"1 ? Detect The Query Object Utilizing A Known Detect The Query Object Utilizing An Object Class Neural Network 206 Unknown Object Class Neural Network 208 Region Proposal Object Classifier Region Proposal Concept Model Model Model Embedding Model Provide The Detected Query Object Within The Image 210 Mggard Fig. 2

14. 发明专利

AU2020202602A1 CLASSIFYING COLORS OF OBJECTS IN DIGITAL IMAGES 未知
公开(公告)号：AU2020202602A1
公开(公告)日：2021-02-11
申请号：AU2020202602
申请日：2020-04-17
申请人： ADOBE INC
发明人： LIN ZHE , COHEN SCOTT , DING ZHIHONG , LING MINGYANG
IPC分类号： G06K9/46 , G06F16/583 , G06K9/62 , G06T7/90
摘要： The present disclosure relates to a color classification system that accurately classifies objects in digital images based on color. In particular, in one or more embodiments, the color classification system utilizes a multidimensional color space and one or more color mappings to match objects to colors. Indeed, the color classification system can accurately and efficiently detect the color of an object utilizing one or more color similarity regions generated in the multidimensional color space. Identify A Request For An Object In A Target Color 202 "Gray Hat" Map The Query Color To Multiple Points In Multidimensional Color Space 204 o QC Detect The Query Object In An Image 206 Determine A Color Match Between The Detected Query Object And The Query Color 208 Provide The Image With The Object Matching The Query Color 210 Fig. 2

15. 发明专利

DK3138296T3 VISNING AF DATA I FORBINDELSE MED ET PROGRAM, DER ER BASERET PÅ AUTOMATISK GENKENDELSE 未知
公开(公告)号：DK3138296T3
公开(公告)日：2020-08-31
申请号：DK15727099
申请日：2015-04-29
申请人： NETFLIX INC
发明人： KANSARA APURVAKUMAR DILIPKUMAR , GARCIA-SHELTON TUSSANEE , HUANG SHIHCHI , WU CHRISTINE SUEJANE
IPC分类号： H04N21/234 , G06F16/583 , G06F16/783 , G11B27/034 , G11B27/10 , G11B27/11 , G11B27/28 , H04N21/233 , H04N21/41 , H04N21/43 , H04N21/81 , H04N21/8547

16. 发明专利

ES2771106T3 Transferencia térmica de imagen compuesta con marca legible mediante escaneo 未知
公开(公告)号：ES2771106T3
公开(公告)日：2020-07-06
申请号：ES14825021
申请日：2014-12-15
申请人： AVERY DENNISON RETAIL INFORMATION SERVICES LLC
发明人： DE CASTRO PAULO
IPC分类号： B41M3/12 , B32B37/00 , B32B37/06 , B41M5/025 , B41M5/035 , B42D25/305 , B44C1/17 , D06P5/00 , D06P5/24 , D06Q1/12 , G06F16/583 , G06Q50/18
摘要： Una transferencia térmica de imagen compuesta (100), que comprende: una transferencia térmica que tiene una marca legible mediante escaneo (102) embebida en la transferencia térmica; la marca legible mediante escaneo (102) colocada en la transferencia térmica en un área de una imagen; un artículo de ropa adecuado para recibir la transferencia térmica, conteniendo la marca legible mediante escaneo (102) información relacionada con la imagen, en la que la marca legible mediante escaneo (102) se enlaza a una base de datos que tiene información relacionada con un tema, caracterizado por que la marca legible mediante escaneo comprende una marca de agua que no es visible para un usuario.

17. 发明专利

IL237548A Methods for categorizing input images for use e.g. as a gateway to authentication systems 未知
公开(公告)号：IL237548A
公开(公告)日：2020-05-31
申请号：IL23754815
申请日：2015-03-04
申请人： AU10TIX LTD
IPC分类号： G06F16/583 , G06K9/00 , G06K9/46 , G06T5/40
摘要： A scanned image authentication system employing at least one image authentication station which is unable to process scanned images which do not belong to at least one individual scanned image category, the system comprising a gateway image processor operative for on-line filtering of an incoming scanned image to determine whether or not the incoming scanned image belongs to the scanned image category, and a computer-implemented interface operative to transmit each individual scanned image to the at least one image authentication station only if the gateway image processor has determined that the individual scanned image belongs to the at least one individual scanned image category.

18. 发明专利

CA2671025C DIGITAL IMAGE ARCHIVING AND RETRIEVAL USING A MOBILE DEVICE SYSTEM 未知
公开(公告)号：CA2671025C
公开(公告)日：2019-07-16
申请号：CA2671025
申请日：2007-11-28
申请人： GOOGLE INC
发明人： CHAUDHURY KRISHNENDU , GARG ASHUTOSH , PHUKAN PRASENJIT , SARAF ARVIND
IPC分类号： G06F16/583 , G06K9/62
摘要： A computer-implemented method of managing information is disclosed. The method can include receiving a message from a mobile device configured to connect to a mobile device network (the message including a digital image taken by the mobile device and including information corresponding to words), determining the words from the digital image information using optical character recognition, indexing the digital image based on the words, and storing the digital image for later retrieval of the digital image based on one or more received search terms.

19. 发明专利

BR112022017004A2 EXTRAÇÃO DE DADOS ESTRUTURADOS POR APRENDIZADO DE MÁQUINA A PARTIR DE IMAGEM DE DOCUMENTO 未知
公开(公告)号：BR112022017004A2
公开(公告)日：2022-10-11
申请号：BR112022017004
申请日：2021-03-01
申请人： UBER TECHNOLOGIES INC
发明人： HIMAANSHU GUPTA , XUEWEN ZHANG , JINGCHEN LIU , ABI KOMMA , ANUPAM DIKSHIT , MRIDUL GUPTA , ZEJUN HUANG
IPC分类号： G06N20/20 , G06F16/35 , G06F16/583
摘要： EXTRAÇÃO DE DADOS ESTRUTURADOS POR APRENDIZADO DE MÁQUINA A PARTIR DE IMAGEM DE DOCUMENTO. A presente invenção refere-se a aplicativo de transcrição de documento que recebe uma imagem de um documento que compreende dados estruturados. O aplicativo de transcrição de documento executa reconhecimento óptico de caractere na imagem do documento para produzir um bloco de texto. O aplicativo de transcrição de documento aplica o bloco de texto para um primeiro modelo de aprendizado de máquina para determinar um mapa de calor para uma classe de dados nos dados estruturados na imagem do documento. O aplicativo de transcrição de documento aplica a imagem do documento e o mapa de calor para um segundo modelo de aprendizado de máquina para identificar uma região da imagem do documento representando a classe de dados. O aplicativo de transcrição de documento gera, usando a região identificada e o bloco de texto, um arquivo de dados estruturados.

20. 发明专利

ES2924268T3 Procedimiento, aparato y dispositivo electrónico de recuperación de imágenes 未知
公开(公告)号：ES2924268T3
公开(公告)日：2022-10-05
申请号：ES18839135
申请日：2018-07-25
申请人： HANGZHOU HIKVISION DIGITAL TEC
发明人： CHEN CHANGHUAI
IPC分类号： G06F16/583 , G06K9/62
摘要： Las realizaciones de la presente solicitud proporcionan un método, dispositivo y dispositivo electrónico de recuperación de imágenes, en el que el método incluye: adquirir una imagen de consulta; determinar una característica objetivo de la imagen de consulta en base a una red neuronal profunda previamente entrenada; en el que la red neuronal profunda se obtiene entrenando según las respectivas imágenes de muestra y características predeterminadas, que pueden formar la característica objetivo, correspondiente a las respectivas imágenes de muestra; obtener características objetivo de una pluralidad de imágenes a recuperar; calcular una similitud entre la característica objetivo de la imagen de consulta y la característica objetivo de cada imagen a recuperar; determinar una imagen de recuperación correspondiente a la imagen de consulta de la pluralidad de imágenes a recuperar de acuerdo con las similitudes calculadas. El método, el dispositivo y el dispositivo electrónico de recuperación de imágenes proporcionados por las realizaciones de la presente solicitud pueden mejorar la precisión de la recuperación de imágenes. (Traducción automática con Google Translate, sin valor legal)

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