-
31.发明公开
公开(公告)号：EP3926531A1
公开(公告)日：2021-12-22
申请号：EP21179781.6
申请日：2021-06-16
发明人： Kurma, Sai Sree Bhargav , KALRA, Kanika , Vadakkeeveetil Sreelatha, Silpa , Patwardhan, Manasi , Karande, Shirish Subhash
摘要： This disclosure relates generally to visio-linguistic understanding. Conventional methods use contextual visio-linguistic reasoner for visio-linguistic understanding which requires more compute power and large amount of pre-training data. Embodiments of the present disclosure provide a method for visio-linguistic understanding using contextual language model reasoner. The method converts the visual information of an input image into a format that the contextual language model reasoner understands and accepts for a downstream task. The method utilizes the image captions and confidence score associated with the image captions along with a knowledge graph to obtain a combined input in a format compatible with the contextual language model reasoner. Contextual embeddings corresponding to the downstream task is obtained using the combined input. The disclosed method is used to solve several downstream tasks such as scene understanding, visual question answering, visual common-sense reasoning and so on.
-
公开(公告)号：EP2981934B1
公开(公告)日：2021-12-15
申请号：EP14715153.4
申请日：2014-03-05
发明人： BAUM, Kevin L. , ISHTIAQ, Faisal
IPC分类号： G06K9/32
-
-
34.发明公开
公开(公告)号：EP3918513A1
公开(公告)日：2021-12-08
申请号：EP19755725.9
申请日：2019-07-26
申请人： Intuit Inc.
发明人： TORRES, Terrence J.
-
公开(公告)号：EP3915272A1
公开(公告)日：2021-12-01
申请号：EP20744793.9
申请日：2020-01-24
申请人： Gracenote Inc.
IPC分类号： H04N21/81 , H04N21/4728 , H04N21/431 , H04N21/478 , G06K9/00 , G06K9/32
-
公开(公告)号：EP3915271A1
公开(公告)日：2021-12-01
申请号：EP20744498.5
申请日：2020-01-24
申请人： Gracenote Inc.
IPC分类号： H04N21/81 , H04N21/431 , H04N21/478 , H04N21/4728 , G06K9/00 , G06K9/32
-
公开(公告)号：EP3912081A1
公开(公告)日：2021-11-24
申请号：EP19812699.7
申请日：2019-11-14
申请人： Siemens Schweiz AG
发明人： FREY, Christian , ZECHLIN, Oliver
-
公开(公告)号：EP3910548A1
公开(公告)日：2021-11-17
申请号：EP20472005.6
申请日：2020-05-14
申请人： Demax - Holograms AD
发明人： Monovski, Valentin Konstantinov , Laskov, Evgenii Atanassov , Ivanovskii, Andrii , Baldjiev, Anguel Georgiev , Tonchev, Dimitar Antonov , Dinova, Margita Todorova , Dobrev, Mario Roumenov , Akuzov, Danail Trendafilov , Pogan, Ignat , Stoilov, Georgi Ivanov
摘要： The method relates to the security of the documents and goods from forgery. Its application enhances the protective properties of the diffractive optical variable device by improving the inspection procedure, with automatic recognition of machine-readable features added to visually controlled ones. The following operations are performed: after forming the diffractive optical variable element, it is subjected to verification by launching an application using the algorithm on a smartphone with a camera. The user manually positions the smartphone over the optical variable device (1) and shoots it at two substantially different observation conditions. The images are geometrically straightened and correlated with the original one. From the straightened images, color characteristics are found, dynamic effect parameters are measured and hidden information is decrypted. Upon a successful comparison of the characteristics and parameters with predefined values, the originality of the captured diffractive optical variable device is determined.
-
公开(公告)号：EP3906501A1
公开(公告)日：2021-11-10
申请号：EP19853296.2
申请日：2019-12-19
申请人： Zoox, Inc.
发明人： GOEL, Kratarth , TARIQ, Sarah
-
公开(公告)号：EP3905121A1
公开(公告)日：2021-11-03
申请号：EP21170945.6
申请日：2021-04-28
申请人： MOJ.IO Inc.
发明人： Messer, Alan
摘要： A method of operation of a vehicle system includes receiving an image from a visual sensor, identifying an item of interest based on the image, generating a bounding box around the item of interest, categorizing a target object based on the item of interest within the bounding box, calculating a distance based on a width of the bounding box, and communicating the distance for assisting in operation of the vehicle.
-
-
-
-
-
-
-
-
-
搜索结果
1974 条记录 (耗时 0.021 秒)
