中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

95 records (took 0.026 seconds)

    SYSTEM AND METHOD FOR A UNIFIED ARCHITECTURE MULTI-TASK DEEP LEARNING MACHINE FOR OBJECT RECOGNITION
    52.
    发明申请
    SYSTEM AND METHOD FOR A UNIFIED ARCHITECTURE MULTI-TASK DEEP LEARNING MACHINE FOR OBJECT RECOGNITION 审中-公开

    公开(公告)号：US20180307897A1

    公开(公告)日：2018-10-25

    申请号：US16024823

    申请日：2018-06-30

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Mostafa EL-KHAMY ,  Arvind YEDLA ,  Marcel NASSAR ,  Jungwon LEE

    IPC: G06K9/00

    CPC classification number: G06K9/00268 ,  G06K9/00228 ,  G06T7/70 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30201

    Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss. A grouping network forms, based on the first hierarchical-calculated feature, the regression loss for the alignment parameters and the bounding box regression loss, at least one of a box grouping, an alignment parameter grouping, and a non-maximum suppression of the alignment parameters and the detected boxes.

    EFFICIENT POLYPHASE ARCHITECTURE FOR INTERPOLATOR AND DECIMATOR
    53.
    发明申请
    EFFICIENT POLYPHASE ARCHITECTURE FOR INTERPOLATOR AND DECIMATOR 有权

    公开(公告)号：US20180115329A1

    公开(公告)日：2018-04-26

    申请号：US15402651

    申请日：2017-01-10

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Liangbin LI ,  Pranav DAYAL ,  Jungwon LEE ,  Gennady FEYGIN

    IPC: H04B1/00 ,  H04B3/462

    CPC classification number: H04B1/0042 ,  H04B1/0046 ,  H04B3/462

    Abstract: Apparatuses (and methods of manufacturing same), systems, and methods concerning polyphase digital filters are described. In one aspect, an apparatus is provided, including at least one pair of subfilters, each having symmetric coefficients, and a lattice comprising two adders and feedlines corresponding to each of the at least one pair of subfilters, each having symmetric coefficients. In one aspect, the apparatus is a polyphase finite impulse response (FIR) digital filter, including an interpolator and a decimator, where each of the interpolator and the decimator have at least one pair of subfilters, each having symmetric coefficients, and a lattice comprising two adders and feedlines corresponding to each of the at least one pair of subfilters, each having symmetric coefficients.

    METHOD AND APPARATUS FOR NEURAL NETWORK QUANTIZATION
    54.
    发明申请
    METHOD AND APPARATUS FOR NEURAL NETWORK QUANTIZATION 审中-公开

    公开(公告)号：US20180107925A1

    公开(公告)日：2018-04-19

    申请号：US15433531

    申请日：2017-02-15

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yoo Jin CHOI ,  Mostafa EL-KHAMY ,  Jungwon LEE

    IPC: G06N3/08 ,  G06F17/16

    CPC classification number: G06N3/08 ,  G06F17/16 ,  G06N3/063 ,  G06N3/082 ,  G06N3/084

    Abstract: Apparatuses and methods of manufacturing same, systems, and methods for performing network parameter quantization in deep neural networks are described. In one aspect, diagonals of a second-order partial derivative matrix (a Hessian matrix) of a loss function of network parameters of a neural network are determined and then used to weight (Hessian-weighting) the network parameters as part of quantizing the network parameters. In another aspect, the neural network is trained using first and second moment estimates of gradients of the network parameters and then the second moment estimates are used to weight the network parameters as part of quantizing the network parameters. In yet another aspect, network parameter quantization is performed by using an entropy-constrained scalar quantization (ECSQ) iterative algorithm. In yet another aspect, network parameter quantization is performed by quantizing the network parameters of all layers of a deep neural network together at once.

    ELECTRONIC DEVICE, AND ELECTRONIC DEVICE OPERATION METHOD
    58.
    发明公开
    ELECTRONIC DEVICE, AND ELECTRONIC DEVICE OPERATION METHOD 审中-公开

    公开(公告)号：US20230289047A1

    公开(公告)日：2023-09-14

    申请号：US18303180

    申请日：2023-04-19

    Applicant: SAMSUNG ELECTRONICS CO., LTD.

    Inventor： Duyeong CHOI ,  Seonkeun PARK ,  Jungwon LEE ,  Jaewoong CHUNG

    IPC: G06F3/04845 ,  G06F1/16

    CPC classification number: G06F3/04845 ,  G06F1/1652 ,  H04M1/72469

    Abstract: An example may include a display in which a display area of the display is expandable and retractable, a memory, and a processor operatively connected to the display and the memory, wherein the memory includes instructions causing, when executed, the processor to: display an execution screen of a running application on the display area at a first magnification value; expand the display area on the basis of a first designated input; based at least in part on the first designated input, information related to the application, and a first user content included in the execution screen, determine whether to maintain a display magnification of the display at the first magnification value or change the display magnification to a second magnification value different from the first magnification value; and display the execution screen on the expanded display area based on the determined display magnification.

Patent Agency Ranking