一种基于改进结构的级联手势回归算法  

A Cascaded Gesture Regression Algorithm Based on Improved Structure

作　　者：齐嘉锐[1] 李文辉[1] QI Jiarui LI Wenhui(College of Computer Science and Technology, Jilin University, Changchun 130012, China)

机构地区：[1]吉林大学计算机科学与技术学院,长春130012

出　　处：《吉林大学学报（理学版）》2017年第1期89-94,共6页Journal of Jilin University:Science Edition

基　　金：吉林省科技发展计划项目(批准号:20140101181JC)

摘　　要：基于手势特征和回归算法,通过改进传统的3D手势级联回归算法学习效率低的缺陷,提出一种新的基于手势变化特征的手部结构特征算子,有效减少了手部特征算子的维度,并对传统级联回归器进行了适应于手势变化的结构改进,使其针对手部结构的识别效率显著提高.在公开数据库及自建数据库分别进行实验,实验结果表明,该算法在保持手势识别准确度的同时,能有效提高执行效率.Based on the features of gestures and regression algorithms, by improving the defects of the traditional 3D gesture cascading regression algorithm, we proposed a new operator of hand-structure features based on the gestural variations. This operator efficiently reduced the dimension of hand- feature operators. We adapted to the structural improvement o{ traditional cascading regression in terms of the changes of gestures, so that it significantly improved the recognition efficiency aiming at the handy structures. The experiments were carried out on the public databases and self-built databases, and the experimental results indicates that the algorithm can successfully improve the execution efficiency while maintaining the accuracy of gesture recognition.

关 键 词：手势 级联回归 手势结构 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]