Please wait a minute...
FITEE
Front. Inform. Technol. Electron. Eng.  2015, Vol. 16 Issue (11): 917-929    DOI: 10.1631/FITEE.1500080
    
基于鲁棒局部自适应多视角学习的视点无关人体行为识别
Jia-geng Feng, Jun Xiao
Institute of Artificial Intelligence, College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
View-invariant human action recognition via robust locally adaptive multi-view learning
Jia-geng Feng, Jun Xiao
Institute of Artificial Intelligence, College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
 全文: PDF 
摘要： 目的：基于视觉的人体行为识别是一个非常活跃的研究领域。它在智能监控、感知接口和基于内容的视频检索等领域具有广泛的应用前景。然而，一些现实应用场景仍然阻碍行为识别技术的发展，比如现实场景中的动作往往是从任意角度拍摄的。因此与视点无关的行为识别显得十分重要。大量研究者开始致力于行为识别的视点无关性。本文提出一种基于多视角学习的视点无关人体行为识别方法。
创新点：针对现有多视角学习算法在构建近邻图时缺乏数据自适应性的问题，本文提出一种自适应多视角学习算法。此外，还提出一种迭代优化求解方法对所构建的目标函数进行优化求解。
方法：对于单个视角下的所有样本特征数据，构建一个该视角下的L1图。在获得数据的稀疏图结构后，对于单视角下的数据，希望学习一种最优的降维方法，在对原始数据进行降维的同时，最大程度地保持数据内在的局部结构信息；对于不同的视角，取一个非负的权重向量来衡量不同视角的重要程度。对于全部的视角可以统一起来得到目标函数。最后利用迭代优化求解，用支持向量机（SVM）分类。
结论：将本文所提算法应用到视点无关的行为识别中，实验结果表明：该算法能够自适应地选择近邻数与不同特征的权重；与其他几种对比算法相比，本文所提算法的分类准确率更高。
关键词： 视点无关;  行为识别;  多视角学习：L1范数    
Abstract: Human action recognition is currently one of the most active research areas in computer vision. It has been widely used in many applications, such as intelligent surveillance, perceptual interface, and content-based video retrieval. However, some extrinsic factors are barriers for the development of action recognition; e.g., human actions may be observed from arbitrary camera viewpoints in realistic scene. Thus, view-invariant analysis becomes important for action recognition algorithms, and a number of researchers have paid much attention to this issue. In this paper, we present a multi-view learning approach to recognize human actions from different views. As most existing multi-view learning algorithms often suffer from the problem of lacking data adaptiveness in the nearest neighborhood graph construction procedure, a robust locally adaptive multi-view learning algorithm based on learning multiple local L1-graphs is proposed. Moreover, an efficient iterative optimization method is proposed to solve the proposed objective function. Experiments on three public view-invariant action recognition datasets, i.e., ViHASi, IXMAS, and WVU, demonstrate data adaptiveness, effectiveness, and efficiency of our algorithm. More importantly, when the feature dimension is correctly selected (i.e., >60), the proposed algorithm stably outperforms state-of-the-art counterparts and obtains about 6% improvement in recognition accuracy on the three datasets.
Key words: View-invariant    Action recognition    Multi-view learning    L1-norm    Local learning
收稿日期: 2015-03-18 出版日期: 2015-11-04
CLC:  TP391  
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　
Jia-geng Feng
Jun Xiao

引用本文:

Jia-geng Feng, Jun Xiao. View-invariant human action recognition via robust locally adaptive multi-view learning. Front. Inform. Technol. Electron. Eng., 2015, 16(11): 917-929.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/FITEE.1500080    或    http://www.zjujournals.com/xueshu/fitee/CN/Y2015/V16/I11/917

[1] Gopi Ram , Durbadal Mandal , Sakti Prasad Ghoshal , Rajib Kar . 使用猫群算法优化线性天线阵列的最佳阵因子辐射方向图：电磁仿真验证[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(4): 570-577.
[2] Lin-bo Qiao, Bo-feng Zhang, Jin-shu Su, Xi-cheng Lu. 结构化稀疏学习综述[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(4): 445-463.
[3] Rong-Feng Zhang , Ting Deng , Gui-Hong Wang , Jing-Lun Shi , Quan-Sheng Guan . 基于可靠特征点分配算法的鲁棒性跟踪框架[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(4): 545-558.
[4] Yuan-ping Nie, Yi Han, Jiu-ming Huang, Bo Jiao, Ai-ping Li. 基于注意机制编码解码模型的答案选择方法[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(4): 535-544.
[5] Wen-yan Xiao, Ming-wen Wang, Zhen Weng, Li-lin Zhang, Jia-li Zuo. 基于语料库的小学英语认识率及教材选词策略研究[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(3): 362-372.
[6] . 一种基于描述逻辑的体系质量需求建模与验证方法[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(3): 346-361.
[7] Ali Darvish Falehi, Ali Mosallanejad. 使用基于多目标粒子群算法多层自适应模糊推理系统晶闸管控制串联电容器补偿技术的互联多源电力系统动态稳定性增强器[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(3): 394-409.
[8] Li Weigang. 用于评估共同作者学术贡献的第一和其他合作者信用分配模式[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(2): 180-194.
[9] Jun-hong Zhang, Yu Liu. 应用完备集合固有时间尺度分解和混合差分进化和粒子群算法优化的最小二乘支持向量机对柴油机进行故障诊断[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(2): 272-286.
[10] Hui Chen, Bao-gang Wei, Yi-ming Li, Yong-huai Liu, Wen-hao Zhu. 一种易用的实体识别消歧系统评测框架[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(2): 195-205.
[11] Yue-ting Zhuang, Fei Wu, Chun Chen, Yun-he Pan. 挑战与希望：AI2.0时代从大数据到知识[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 3-14.
[12] Bo-hu Li, Hui-yang Qu, Ting-yu Lin, Bao-cun Hou, Xiang Zhai, Guo-qiang Shi, Jun-hua Zhou, Chao Ruan. 基于综合集成研讨厅的群体智能设计研究[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 149-152.
[13] Yong-hong Tian, Xi-lin Chen, Hong-kai Xiong, Hong-liang Li, Li-rong Dai, Jing Chen, Jun-liang Xing, Jing Chen, Xi-hong Wu, Wei-min Hu, Yu Hu, Tie-jun Huang, Wen Gao. AI2.0时代的类人与超人感知：研究综述与趋势展望[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 58-67.
[14] Yu-xin Peng, Wen-wu Zhu, Yao Zhao, Chang-sheng Xu, Qing-ming Huang, Han-qing Lu, Qing-hua Zheng, Tie-jun Huang, Wen Gao. 跨媒体分析与推理：研究进展与发展方向[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 44-57.
[15] Le-kui Zhou, Si-liang Tang, Jun Xiao, Fei Wu, Yue-ting Zhuang. 基于众包标签数据深度学习的命名实体消歧算法[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(1): 97-106.