| 基于Latent SVM的多视角行为识别方法 |
| |
| 引用本文： | 王丹,臧雪柏,陈奋君.基于Latent SVM的多视角行为识别方法[J].吉林大学学报(信息科学版),2016,34(6):747-752. |
| |
| 作者姓名： | 王丹 臧雪柏 陈奋君 |
| |
| 作者单位： | 北华大学信息技术与传媒学院,吉林吉林132011;吉林大学计算机科学与技术学院,长春130012;吉林大学计算机科学与技术学院,长春,130012 |
| |
| 基金项目： | 国家自然科学基金资助项目(61101155)，吉林省发展和改革委员会省级产业创新专项基金资助项目(2016C035) |
| |
| 摘 要： | 为了在静态图像中获取有效信息, 构建行为模型, 提出了行为覆盖区 ACA(Action Coverage Area)和行为核心 AC(Action Core)的概念, 基于 Latent SVM(Support Vector Machine)目标识别方法, 设计了一种多视角行为模型 MVAM(Multiple Viewpoint Action Model)。 建立了独立的用于行为模型训练和测试的行为数据库。 实验表明, 该表示法对静态图像中的人体行为能有效地进行分类和检测。
|
| 关 键 词： | 行为识别 隐变量支持向量机 行为覆盖区 行为核心 多视角行为模型 |
| 收稿时间： | 2016-08-27 |
Multiple-Viewpoint Human Action Recognition Based on Latent SVM |
| |
| WANG Dan,ZANG Xuebai,CHEN Fenjun.Multiple-Viewpoint Human Action Recognition Based on Latent SVM[J].Journal of Jilin University:Information Sci Ed,2016,34(6):747-752. |
| |
| Authors: | WANG Dan ZANG Xuebai CHEN Fenjun |
| |
| Institution: | 1. College of Information Technology, Beihua University, Jilin 132011, China;
2. Department of Computer Science and Technology, Jilin University, Changchun 130012, China |
| |
| Abstract: | We study the concepts of ACA (Action Coverage Area) and AC (Action Core), and design a multiple-viewpoint action model based on the advanced object recogniton method Latent SVM ( Support Vector Machine). Multi-viewpoint ACA can succinctly describe the non-rigid change and appearance variation due to different viewpoints and multi-viewpoint AC, because the auxiliary of ACA can help to improve the efficiency and robustness of action recognition. We created an independent dataset of human action for model training and testing. Experiments showed that our model had high performance in classifying and detecting human action in still images. |
| |
| Keywords: | action recognition latent support vector machine (SVM) action coverage area (ACA) action core(AC) multiple viewpoint action model (MVAM) |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《吉林大学学报(信息科学版)》浏览原始摘要信息 |
| 点击此处可从《吉林大学学报(信息科学版)》下载免费的PDF全文 |
