中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

70 records (took 0.016 seconds)

    Object recognition and description using multimodal recurrent neural network
    1.
    发明授权
    Object recognition and description using multimodal recurrent neural network 有权

    公开(公告)号：US10970603B2

    公开(公告)日：2021-04-06

    申请号：US16205768

    申请日：2018-11-30

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Hongfei Li ,  Jinfeng Yi ,  Jing Xia

    IPC: G06K9/62 ,  G06N3/08

    Abstract: An embodiment of the invention may include a method, computer program product and computer system for image identification and classification. The method, computer program product and computer system may include a computing device which may receive one or more images of a first object from at least two angles linguistic data associated with the first object. The computing device may input the one or more images of the first object into one or more first neural networks and the linguistic data of the first object into one or more second neural networks. The computing device may combine the output of the one or more first neural networks and the one or more second neural networks and generate an identification model based on the combined output of the one or more first neural networks and the one or more second neural networks.

    OBJECT RECOGNITION AND DESCRIPTION USING MULTIMODAL RECURRENT NEURAL NETWORK
    2.
    发明申请
    OBJECT RECOGNITION AND DESCRIPTION USING MULTIMODAL RECURRENT NEURAL NETWORK 审中-公开

    公开(公告)号：US20200175344A1

    公开(公告)日：2020-06-04

    申请号：US16205768

    申请日：2018-11-30

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Hongfei Li ,  Jinfeng Yi ,  Jing Xia

    IPC: G06K9/62 ,  G06N3/08

    Abstract: An embodiment of the invention may include a method, computer program product and computer system for image identification and classification. The method, computer program product and computer system may include a computing device which may receive one or more images of a first object from at least two angles linguistic data associated with the first object. The computing device may input the one or more images of the first object into one or more first neural networks and the linguistic data of the first object into one or more second neural networks. The computing device may combine the output of the one or more first neural networks and the one or more second neural networks and generate an identification model based on the combined output of the one or more first neural networks and the one or more second neural networks.

    INTRADAY CASH FLOW OPTIMIZATION
    5.
    发明申请
    INTRADAY CASH FLOW OPTIMIZATION 审中-公开
    INTRADAY现货流优化

    公开(公告)号：US20150081483A1

    公开(公告)日：2015-03-19

    申请号：US14155760

    申请日：2014-01-15

    Applicant: International Business Machines Corporation

    Inventor： JoAnn P. Brereton ,  Arun Hampapur ,  Hongfei Li ,  Robin Lougee ,  Buyue Qian

    IPC: G06Q40/00

    CPC classification number: G06Q40/12 ,  G06Q40/00 ,  G06Q40/02

    Abstract: Embodiments relate to intraday cash flow optimization. Transactions are accessed on a business-to-business integration network from a plurality of sources linked with payment delivery system data from a financial service system. The transactions are associated with two or more compartmentalized entities. The transactions are characterizes based on the payment delivery system data and an analysis of customer profile data. The transactions associated with two or more compartmentalized entities are linked as integrated information based on the characterizing of the transactions. An intraday receivables prediction engine and an intraday payables prediction engine are applied to the integrated information to produce an estimation of intraday cash flow. The estimation of intraday cash flow is monitored relative to intraday operations optimization conditions. An alert is generated based on determining that at least one of the intraday operations optimization conditions is met.

    Abstract translation: 实施例涉及日内现金流优化。 在与来自金融服务系统的付款传送系统数据相关联的多个源的企业对企业集成网络上访问交易。 这些交易与两个或更多分隔的实体相关联。 这些交易的特点是基于付款系统数据和客户资料数据分析。 与两个或多个分区实体相关联的交易根据交易的特征被链接为综合信息。 将日内应收账款预测引擎和日内应付账款预测引擎应用于综合信息以产生日内现金流量的估计。 相对于日内业务优化条件监测日内现金流量的估计。 基于确定满足日内操作优化条件中的至少一个，生成警报。

Patent Agency Ranking