WO2019229524A3 - 神经网络计算方法和系统及相应的双神经网络实现 - Google Patents

神经网络计算方法和系统及相应的双神经网络实现 Download PDF

Info

Publication number
WO2019229524A3
WO2019229524A3 PCT/IB2019/000603 IB2019000603W WO2019229524A3 WO 2019229524 A3 WO2019229524 A3 WO 2019229524A3 IB 2019000603 W IB2019000603 W IB 2019000603W WO 2019229524 A3 WO2019229524 A3 WO 2019229524A3
Authority
WO
WIPO (PCT)
Prior art keywords
neural network
feature
calculation method
input
regard
Prior art date
Application number
PCT/IB2019/000603
Other languages
English (en)
French (fr)
Other versions
WO2019229524A2 (zh
Inventor
刘一楠
Original Assignee
赛灵思公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 赛灵思公司 filed Critical 赛灵思公司
Publication of WO2019229524A2 publication Critical patent/WO2019229524A2/zh
Publication of WO2019229524A3 publication Critical patent/WO2019229524A3/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N99/00—Subject matter not provided for in other groups of this subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)

Abstract

提出了一种神经网络计算方法、系统及相应双网络实现。所述计算方法，包括：获取神经网络计算的输入，所述输入包括两个或以上类别的特征；基于第一类别神经网络提取所述输入针对所述第一类别的第一特征;基于第二类别参考神经网络提取所述输入针对所述第二类别的第二参考特征；将所述第二参考特征引入所述第一特征以去除所述第一特征中第二特征的影响；以及基于叠加了所述第二参考特征的所述第一特征完成针对所述第一特征的神经网络分类计算。本发明通过在将经由神经网络求取的特征向量送入分类器之前，与相关参考特征向量相叠加来去除相关特征对目标特征的分布影响，由此提升分类准确率。
PCT/IB2019/000603 2018-05-31 2019-05-30 神经网络计算方法和系统及相应的双神经网络实现 WO2019229524A2 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810550516.1A CN110555340B (zh) 2018-05-31 2018-05-31 神经网络计算方法和系统及相应的双神经网络实现
CN201810550516.1 2018-05-31

Publications (2)

Publication Number Publication Date
WO2019229524A2 WO2019229524A2 (zh) 2019-12-05
WO2019229524A3 true WO2019229524A3 (zh) 2020-05-22

Family

ID=68697424

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2019/000603 WO2019229524A2 (zh) 2018-05-31 2019-05-30 神经网络计算方法和系统及相应的双神经网络实现

Country Status (2)

Country Link
CN (1) CN110555340B (zh)
WO (1) WO2019229524A2 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111723691B (zh) * 2020-06-03 2023-10-17 合肥的卢深视科技有限公司 一种三维人脸识别方法、装置、电子设备及存储介质
CN112257526B (zh) * 2020-10-10 2023-06-20 中国科学院深圳先进技术研究院 一种基于特征交互学习的动作识别方法及终端设备
CN112395971A (zh) * 2020-11-16 2021-02-23 公安部第三研究所 基于StarGAN的不同量化光照及角度条件下人脸测试图像生成方法、应用及存储介质
CN112766215A (zh) * 2021-01-29 2021-05-07 北京字跳网络技术有限公司 人脸融合方法、装置、电子设备及存储介质
CN115034312B (zh) * 2022-06-14 2023-01-06 燕山大学 一种双神经网络模型卫星电源系统故障诊断方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130301885A1 (en) * 2003-07-18 2013-11-14 Canon Kabushiki Kaisha Image processing device, imaging device, image processing method
US9471886B2 (en) * 2013-10-29 2016-10-18 Raytheon Bbn Technologies Corp. Class discriminative feature transformation
CN107403200A (zh) * 2017-08-10 2017-11-28 北京亚鸿世纪科技发展有限公司 改进图像分割算法结合深度学习的多重不良图片分类方法
US20170351952A1 (en) * 2016-06-01 2017-12-07 Kla-Tencor Corporation Systems and methods incorporating a neural network and a forward physical model for semiconductor applications

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101719278B1 (ko) * 2015-04-14 2017-04-04 (주)한국플랫폼서비스기술 비주얼 콘텐츠기반 영상 인식을 위한 딥러닝 프레임워크 및 영상 인식 방법
CN107145857B (zh) * 2017-04-29 2021-05-04 深圳市深网视界科技有限公司 人脸属性识别方法、装置和模型建立方法
CN107766850B (zh) * 2017-11-30 2020-12-29 电子科技大学 基于结合人脸属性信息的人脸识别方法
CN107895160A (zh) * 2017-12-21 2018-04-10 曙光信息产业(北京)有限公司 人脸检测与识别装置及方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130301885A1 (en) * 2003-07-18 2013-11-14 Canon Kabushiki Kaisha Image processing device, imaging device, image processing method
US9471886B2 (en) * 2013-10-29 2016-10-18 Raytheon Bbn Technologies Corp. Class discriminative feature transformation
US20170351952A1 (en) * 2016-06-01 2017-12-07 Kla-Tencor Corporation Systems and methods incorporating a neural network and a forward physical model for semiconductor applications
CN107403200A (zh) * 2017-08-10 2017-11-28 北京亚鸿世纪科技发展有限公司 改进图像分割算法结合深度学习的多重不良图片分类方法

Also Published As

Publication number Publication date
WO2019229524A2 (zh) 2019-12-05
CN110555340B (zh) 2022-10-18
CN110555340A (zh) 2019-12-10

Similar Documents

Publication Publication Date Title
WO2019229524A3 (zh) 神经网络计算方法和系统及相应的双神经网络实现
Sirunyan et al. Combined search for electroweak production of charginos and neutralinos in proton-proton collisions at $$\sqrt {s}= 13$$ TeV
WO2019040866A3 (en) APPARATUS AND METHOD FOR COLLECTING AGRICULTURAL DATA AND AGRICULTURAL OPERATIONS
WO2019204632A8 (en) Method and system for rapid genetic analysis
WO2018125339A3 (en) Computer vision systems and methods for detecting and modeling features of structures in images
WO2013148519A3 (en) Range-cued object segmentation system and method
EP2150053B8 (en) Vehicle periphery monitoring device, vehicle periphery monitoring program and vehicle periphery monitoring method
EP2763077A3 (en) Method and apparatus for sensor aided extraction of spatio-temporal features
DE602007005833D1 (de) Sprachaktivitätdetektionssystem und verfahren
WO2018114047A8 (de) Lösungsmittel sowie verfahren zum lösen eines kunststoffs von einem feststoff innerhalb einer suspension
WO2018197862A3 (en) DETERMINATION OF RISK POSED BY VEGETATION
WO2017097596A3 (de) Verfahren und steuergerät zum erkennen einer möglichen kollision eines unbemannten luftfahrzeugs mit einem objekt
WO2016036637A3 (en) Generating metadata for audio object
EP3547210A3 (en) Decentralized video tracking
WO2020076644A3 (en) Methods for automated detection of cervical pre-cancers with a low-cost, point-of-care, pocket colposcope
EP3032457A3 (en) Driver assistance for a vehicle
WO2020197233A3 (ko) 다중시기 측부혈류 영상을 생성하기 위한 학습 방법 및 기계 학습을 이용한 다중시기 측부혈류 영상 생성 방법
WO2020182690A3 (en) Method for creating a collision detection training set including ego part exclusion
MX2020000433A (es) Metodo para recuperar acido fosforico del lecho de residuos de fermentacion o residuos de fermentacion y reutilizacion del mismo.
KR101881505B1 (ko) 지문 위조 판별 모델 생성 방법 및 이를 이용한 지문 위조 판별 방법
WO2020061605A3 (de) Verfahren zum fokussieren einer kamera
WO2019106437A3 (en) Matching bids for work with offers for work
WO2020256517A3 (ko) 전방위 화상정보 기반의 자동위상 매핑 처리 방법 및 그 시스템
SE1750592A1 (en) Method and system for controlling a formation of vehicles during drive of said vehicles
WO2020178815A3 (en) System and method for adjustable production line inspection

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19811873

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19811873

Country of ref document: EP

Kind code of ref document: A2