首页> 外国专利> HUMAN FACE DETECTION METHOD AND HUMAN FACE DETECTION SYSTEM

HUMAN FACE DETECTION METHOD AND HUMAN FACE DETECTION SYSTEM

机译:人脸检测方法和人脸检测系统

摘要

A human face detection method and a human face detection system. The human face detection method comprises the following steps: establishing a convolutional neural network framework, the convolutional neural network framework at least comprising a candidate region generating network, a correction network and a multi-information output network; connecting the convolutional neural network framework to a data preparation module; running the candidate region generating network, and generating multiple first human face candidate region frames; running the correction network, screening the first human face candidate region frames, correcting the positions of residual first human face candidate region frames, and performing non-maximum suppression; and running the multi-information output network, screening the first human face candidate region frames, correcting the positions of the residual first human face candidate region frames, performing non-maximum suppression, and outputting human face feature points and human facial poses corresponding to the residual first human face candidate region frames and the first human face candidate region frames.
机译:人脸检测方法和人脸检测系统。人脸检测方法包括以下步骤:建立卷积神经网络框架,该卷积神经网络框架至少包括候选区域生成网络,校正网络和多信息输出网络。将卷积神经网络框架连接到数据准备模块;运行候选区域生成网络,并生成多个第一人脸候选区域帧;运行校正网络,筛选第一人脸候选区域框,校正剩余的第一人脸候选区域框的位置,并进行非最大抑制。运行多信息输出网络,筛选第一人脸候选区域框,校正剩余的第一人脸候选区域框的位置,进行非最大抑制,并输出与该人脸对应的人脸特征点和人脸姿势。剩余的第一人脸候选区域框和第一人脸候选区域框。

著录项

  • 公开/公告号WO2019091271A1

    专利类型

  • 公开/公告日2019-05-16

    原文格式PDF

  • 申请/专利权人 SUZHOU KEDA TECHNOLOGY CO. LTD.;

    申请/专利号WO2018CN110703

  • 发明设计人 JIN ZHAOLONG;ZHAO BO;CHEN WEIDONG;

    申请日2018-10-17

  • 分类号G06K9;G06N3/04;

  • 国家 WO

  • 入库时间 2022-08-21 11:54:52

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号