知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

6 records (took 0.028 seconds)

    System for training embedding network
    2.
    发明授权
    System for training embedding network 有权

    公开(公告)号：US11823488B1

    公开(公告)日：2023-11-21

    申请号：US17215762

    申请日：2021-03-29

    Applicant: AMAZON TECHNOLOGIES, INC.

    Inventor： Rajeev Ranjan ,  Manoj Aggarwal ,  Gerard Guy Medioni

    IPC: G06V40/12 ,  G06N3/08 ,  G06V40/13

    CPC classification number: G06V40/1365 ,  G06N3/08 ,  G06V40/1318

    Abstract: Biometric input, such as an image of a hand, may be processed to determine embedding vector data that may be used to identify users. Accuracy of the identification is improved by using high resolution inputs to a deep convolutional neural network (DCNN) that is trained to generate the embedding vector data that is representative of features in the input. Training data sets are expensive to develop and thus may be relatively small. During training of the DCNN, confidence loss values corresponding to the entire input as well as particular patches or portions of the input are determined. These patch-wise confidence loss values mitigate potential overfitting during training of the DCNN and improve overall performance of the trained DCNN to determine embedding vector data suitable for identification.

    ELECTRONIC DEVICE FOR AUTOMATED USER IDENTIFICATION
    3.
    发明申请
    ELECTRONIC DEVICE FOR AUTOMATED USER IDENTIFICATION 有权

    公开(公告)号：US20210097547A1

    公开(公告)日：2021-04-01

    申请号：US16585328

    申请日：2019-09-27

    Applicant: Amazon Technologies, Inc.

    Inventor： Nichole Stockman ,  Korwin Jon Smith ,  Douglas Andrew Hungarter ,  Joshua Adam Cowan ,  Jared Corso ,  Rajeev Ranjan ,  Prithviraj Banerjee ,  Matthew Christopher Smith

    IPC: G06Q20/40 ,  G06K9/00

    Abstract: This disclosure describes techniques for providing instructions when receiving biometric data associated with a user. For instance, a user-recognition device may detect a portion of a user, such as a hand. The user-recognition device may then display a first graphical element indicating a target location for placing the portion of the user above the user-recognition device. Additionally, the user-recognition device may determine locations of the portion of the user above the user-recognition device. The user-recognition device may then display a second graphical element indicating the locations, such as when the locations are not proximate to the target location. Additionally, the user-recognition device may display instructions for moving the portion of the user to the target location. Based on detecting that the location of the portion of the user is proximate to the target location, the user-recognition device may send data representing the portion of the user to a remote system.

    System using multimodal decorrelated embedding model
    4.
    发明授权
    System using multimodal decorrelated embedding model 有权

    公开(公告)号：US11688198B1

    公开(公告)日：2023-06-27

    申请号：US17457551

    申请日：2021-12-03

    Applicant: AMAZON TECHNOLOGIES, INC.

    Inventor： Rajeev Ranjan ,  Gerard Guy Medioni ,  Manoj Aggarwal ,  Dilip Kumar

    IPC: G06V40/13 ,  G06V10/94 ,  G06F18/213 ,  G06F18/214

    CPC classification number: G06V40/1318 ,  G06F18/213 ,  G06F18/2148 ,  G06V10/95

    Abstract: A biometric identification system uses inputs acquired using different modalities. A model having an intersection branch and an XOR branch is trained to determine an embedding using features present in all modalities (an intersection of modalities), and features that are distinctive to each modality (an XOR of that modality relative to the other modality(s)). During training, a first loss function is used to determine a first loss value with respect to the branches. Probability distributions are determined for the output from the branches, corresponding to the intersection and XORs of each modality. A second loss function uses these probability distributions to determine a second loss value. A total loss function for training the model may be a sum of the first loss and the second loss. Once trained, the model may process query inputs to determine embedding data for comparison with embedding data of a previously enrolled user.

    System for training neural network using ordered classes
    6.
    发明授权
    System for training neural network using ordered classes 有权

    公开(公告)号：US11868443B1

    公开(公告)日：2024-01-09

    申请号：US17302770

    申请日：2021-05-12

    Applicant: AMAZON TECHNOLOGIES, INC.

    Inventor： Rajeev Ranjan ,  Prithviraj Banerjee ,  Manoj Aggarwal ,  Gerard Guy Medioni ,  Dilip Kumar

    IPC: G06F18/2431 ,  G06N3/08 ,  G06F18/214 ,  G06F18/2113 ,  G06N3/045

    CPC classification number: G06F18/2431 ,  G06F18/214 ,  G06F18/2113 ,  G06N3/045 ,  G06N3/08

    Abstract: A neural network is trained to process input data and generate a classification value that characterizes the input with respect to an ordered continuum of classes. For example, the input data may comprise an image and the classification value may be indicative of a quality of the image. The ordered continuum of classes may represent classes of quality of the image ranging from “worst”, “bad”, “normal”, “good”, to “best”. During training, loss values are determined using an ordered classification loss function. The ordered classification loss function maintains monotonicity in the loss values that corresponds to placement in the continuum. For example, the classification value for a “bad” image will be less than the classification value indicative of a “best” image. The classification value may be used for subsequent processing. For example, biometric input data may be required to have a minimum classification value for further processing.

Patent Agency Ranking