利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

245 条记录 (耗时 0.022 秒)

    SYNCHRONIZING A HIGH-SPEED SIGNALING INTERCONNECT
    2.
    发明公开
    SYNCHRONIZING A HIGH-SPEED SIGNALING INTERCONNECT 审中-公开

    公开(公告)号：EP4135231A1

    公开(公告)日：2023-02-15

    申请号：EP22190301.6

    申请日：2022-08-12

    申请人： Nvidia Corporation

    发明人： KRISHNAMURTHY, Adithya Hrudhayan ,  CHADHA, Ish

    IPC分类号： H04L1/00 ,  H04L27/26

    摘要： A system includes a first device and a second device coupled to a link. The first device is to transmit one or more request frames for synchronization of a data layer, each request frame including a quantity of bits and an error code. The second device is to receive a first set of bits corresponding to the quantity of bits in each request frame. The second device is to perform an error decode operation on the first set of bits using a first portion of the first set of bits and determine the first set of bits correspond to a frame boundary of the one more request frames responsive to a success of the error decode operation. The second device is to transmit an acknowledgement of the synchronization of the data layer based on determining the first set of bits corresponds to the frame boundary.

    DATA COMPRESSION FOR A NEURAL NETWORK
    5.
    发明公开
    DATA COMPRESSION FOR A NEURAL NETWORK 审中-公开

    公开(公告)号：EP3713093A1

    公开(公告)日：2020-09-23

    申请号：EP20161795.8

    申请日：2020-03-09

    申请人： NVIDIA Corporation

    发明人： Garcia Garcia, David Rigel

    IPC分类号： H03M7/30 ,  G06F7/483 ,  G06F7/544 ,  G06N3/063 ,  H03M7/24

    摘要： Systems and methods for generating a representative value of a data set by first compressing a portion of values in the data set to determine a first common value and further compressing a subset of the portion of values to determine a second common value. The representative value is generated by taking the difference between the first common value and the second common value, wherein the representative value corresponds to a mathematical relationship between the first and second common values and each value within the subset of the portion of values. The representative value requires less storage than the first and second common values.

    SEMANTIC IMAGE SYNTHESIS FOR GENERATING SUBSTANTIALLY PHOTOREALISTIC IMAGES USING NEURAL NETWORKS
    6.
    发明公开
    SEMANTIC IMAGE SYNTHESIS FOR GENERATING SUBSTANTIALLY PHOTOREALISTIC IMAGES USING NEURAL NETWORKS 审中-公开

    公开(公告)号：EP3686848A1

    公开(公告)日：2020-07-29

    申请号：EP19217022.3

    申请日：2019-12-17

    申请人： Nvidia Corporation

    发明人： PARK, Taesung ,  LIU, Ming-Yu ,  WANG, Ting-Chun ,  ZHU, Junyan

    IPC分类号： G06T11/00 ,  G06T11/60

    摘要： A user can create a basic semantic layout that includes two or more regions identified by the user, each region being associated with a semantic label indicating a type of object(s) to be rendered in that region. The semantic layout can be provided as input to an image synthesis network. The network can be a trained machine learning network, such as a generative adversarial network (GAN), that includes a conditional, spatially-adaptive normalization layer for propagating semantic information from the semantic layout to other layers of the network. The synthesis can involve both normalization and de-normalization, where each region of the layout can utilize different normalization parameter values. An image is inferred from the network, and rendered for display to the user. The user can change labels or regions in order to cause a new or updated image to be generated.

    MESH RECONSTRUCTION USING DATA-DRIVEN PRIORS
    7.
    发明公开
    MESH RECONSTRUCTION USING DATA-DRIVEN PRIORS 审中-公开

    公开(公告)号：EP3671639A1

    公开(公告)日：2020-06-24

    申请号：EP19211501.2

    申请日：2019-11-26

    申请人： Nvidia Corporation

    发明人： GALLO, Orazio ,  BADKI, Abhishek

    IPC分类号： G06T7/55

    摘要： One embodiment of a method includes predicting one or more three-dimensional (3D) mesh representations based on a plurality of digital images, wherein the one or more 3D mesh representations are refined by minimizing at least one difference between the one or more 3D mesh representations and the plurality of digital images.