首页> 外国专利> APPARATUS AND METHOD FOR VIDEO SENSOR-BASED HUMAN ACTIVITY AND FACIAL EXPRESSION MODELING AND RECOGNITION

APPARATUS AND METHOD FOR VIDEO SENSOR-BASED HUMAN ACTIVITY AND FACIAL EXPRESSION MODELING AND RECOGNITION

机译:基于视频传感器的人类活动和面部表情建模与识别的装置和方法

摘要

The present invention relates to modeling and recognition of the body motion and facial expression, human motion process to obtain as the principle element of the depth and / or a binary-like image of the body motion is extracted from the video clip in the modeling method and the principle elements acquiring the ICA representations based on and the ICA process to obtain a codebook using the process and the vector quantization to determine a feature through the LDA based on the expression and the process and the conditions for determining the observed symbol sequence from the video clip transition and divide that by using the HMM is based on the observation matrix includes the step of learning the human body operates the operation part or expression as a human body or the face has the advantage it is possible to recognize, without (segmenting). ; human activity recognition, facial expression, HMM, LDA.
机译:本发明涉及人体运动和面部表情的建模和识别,以建模方法从视频剪辑中提取人体运动的深度和/或二进制图像作为人体运动的过程的主要元素。原理要素基于ICA表示和ICA进程获取密码簿,该进程使用进程和向量量化通过LDA根据表达式和进程以及通过条件确定观测符号序列的条件通过LDA确定特征。视频剪辑的过渡和划分(通过使用HMM基于观察矩阵)包括以下步骤:学习人体对人体或面部的操作部分或表情的操作方式,从而可以识别而无需(分段) 。 ;人类活动识别,面部表情,HMM,LDA。

著录项

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