知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

112 records (took 0.023 seconds)

    LOSS-ERROR-AWARE QUANTIZATION OF A LOW-BIT NEURAL NETWORK
    102.
    发明申请
    LOSS-ERROR-AWARE QUANTIZATION OF A LOW-BIT NEURAL NETWORK 有权

    公开(公告)号：US20210019630A1

    公开(公告)日：2021-01-21

    申请号：US16982441

    申请日：2018-07-26

    Applicant: Anbang YAO ,  Aojun ZHOU ,  Kuan WANG ,  Hao ZHAO ,  Yurong CHEN ,  Intel Corporation

    Inventor： Anbang Yao ,  Aojun Zhou ,  Kuan Wang ,  Hao Zhao ,  Yurong Chen

    IPC: G06N3/08 ,  G06K9/62 ,  G06N3/04

    Abstract: Methods, apparatus, systems and articles of manufacture for loss-error-aware quantization of a low-bit neural network are disclosed. An example apparatus includes a network weight partitioner to partition unquantized network weights of a first network model into a first group to be quantized and a second group to be retrained. The example apparatus includes a loss calculator to process network weights to calculate a first loss. The example apparatus includes a weight quantizer to quantize the first group of network weights to generate low-bit second network weights. In the example apparatus, the loss calculator is to determine a difference between the first loss and a second loss. The example apparatus includes a weight updater to update the second group of network weights based on the difference. The example apparatus includes a network model deployer to deploy a low-bit network model including the low-bit second network weights.

    Recognition of activity in a video image sequence using depth information
    104.
    发明授权
    Recognition of activity in a video image sequence using depth information 有权

    公开(公告)号：US10860844B2

    公开(公告)日：2020-12-08

    申请号：US16098648

    申请日：2016-06-02

    Applicant: INTEL CORPORATION

    Inventor： Shaopeng Tang ,  Anbang Yao ,  Yurong Chen

    IPC: G06K9/00 ,  G06K9/46 ,  G06K9/62 ,  G06T7/215 ,  G06T7/579 ,  G06T7/593

    Abstract: Techniques are provided for recognition of activity in a sequence of video image frames that include depth information. A methodology embodying the techniques includes segmenting each of the received image frames into a multiple windows and generating spatio-temporal image cells from groupings of windows from a selected sub-sequence of the frames. The method also includes calculating a four dimensional (4D) optical flow vector for each of the pixels of each of the image cells and calculating a three dimensional (3D) angular representation from each of the optical flow vectors. The method further includes generating a classification feature for each of the image cells based on a histogram of the 3D angular representations of the pixels in that image cell. The classification features are then provided to a recognition classifier configured to recognize the type of activity depicted in the video sequence, based on the generated classification features.

    METHODS AND SYSTEMS FOR BOOSTING DEEP NEURAL NETWORKS FOR DEEP LEARNING
    108.
    发明申请
    METHODS AND SYSTEMS FOR BOOSTING DEEP NEURAL NETWORKS FOR DEEP LEARNING 审中-公开

    公开(公告)号：US20200026999A1

    公开(公告)日：2020-01-23

    申请号：US16475076

    申请日：2017-04-07

    Applicant: INTEL CORPORATION

    Inventor： Libin Wang ,  Yiwen Guo ,  Anbang Yao ,  Dongqi Cai ,  Lin Xu ,  Ping Hu ,  Shangong Wang ,  Wenhua Cheng ,  Yurong Chen

    IPC: G06N3/08 ,  G06N3/04 ,  G06N3/063 ,  G06T1/20 ,  G06K9/62

    Abstract: Methods and systems are disclosed for boosting deep neural networks for deep learning. In one example, in a deep neural network including a first shallow network and a second shallow network, a first training sample is processed by the first shallow network using equal weights. A loss for the first shallow network is determined based on the processed training sample using equal weights. Weights for the second shallow network are adjusted based on the determined loss for the first shallow network. A second training sample is processed by the second shallow network using the adjusted weights. In another example, in a deep neural network including a first weak network and a second weak network, a first subset of training samples is processed by the first weak network using initialized weights. A classification error for the first weak network on the first subset of training samples is determined. The second weak network is boosted using the determined classification error of the first weak network with adjusted weights. A second subset of training samples is processed by the second weak network using the adjusted weights.

    DYNAMIC EMOTION RECOGNITION IN UNCONSTRAINED SCENARIOS
    109.
    发明申请
    DYNAMIC EMOTION RECOGNITION IN UNCONSTRAINED SCENARIOS 审中-公开

    公开(公告)号：US20190325203A1

    公开(公告)日：2019-10-24

    申请号：US16471106

    申请日：2017-01-20

    Applicant: INTEL CORPORATION

    Inventor： Anbang Yao ,  Dongqi Cai ,  Ping Hu ,  Shandong Wang ,  Yurong Chen

    IPC: G06K9/00 ,  G06N3/08 ,  G06N3/04 ,  G06N20/00 ,  G06K9/46

    Abstract: An apparatus for dynamic emotion recognition in unconstrained scenarios is described herein. The apparatus comprises a controller to pre-process image data and a phase-convolution mechanism to build lower levels of a CNN such that the filters form pairs in phase. The apparatus also comprises a phase-residual mechanism configured to build middle layers of the CNN via plurality of residual functions and an inception-residual mechanism to build top layers of the CNN by introducing multi-scale feature extraction. Further, the apparatus comprises a fully connected mechanism to classify extracted features.

    Human detection in high density crowds
    110.
    发明授权
    Human detection in high density crowds 有权

    公开(公告)号：US10402633B2

    公开(公告)日：2019-09-03

    申请号：US15521512

    申请日：2016-05-23

    Applicant: Intel Corporation

    Inventor： Shaopeng Tang ,  Yurong Chen

    IPC: G06K9/00 ,  G06T7/11 ,  H04N7/18 ,  H04N13/204

    Abstract: The present disclosure describes a non-learning based process and apparatus for detecting humans in an image. This may include receiving an image that has pixel distance information from a camera and using that to determine a height of the pixel above a ground surface. One or more regions may then be identified that may include a head and shoulders of an individual in the image. A multiple threshold technique may be used to remove some background regions, and a mean-shift technique used to find the local highest regions that may be combination of head and shoulders of the person. In embodiments, the view angle and/or the height of the camera may not be fixed.

Patent Agency Ranking