Hierarchical Motion Evolution for Action Recognition

CASIA OpenIR > 智能感知与计算研究中心

	Hierarchical Motion Evolution for Action Recognition
	Wang Hongsong(王洪松)1,2,3 ; Wang Wei(王威)1,2,3 ; Wang Liang(王亮)1,2,3
	2015
会议名称	Asian Conference on Pattern Recognition（ACPR）
会议日期	2015-11-01
会议地点	Kuala Lumpur, Malaysia
摘要	Human action can be decomposed into a series of temporally correlated motions. Since the traditional bag-of-words framework based on local features cannot model global motion evolution of actions, models like Recurrent Neural Network (RNN) and VideoDarwin are accordingly explored to capture video-wise temporal information. Inspired by VideoDarwin, in this paper, we present a novel hierarchical scheme to learn better video representation, called HiVideoDarwin. Specifically, we first use different ranking machines to learn motion descriptors of local video clips. Then, in order to model motion evolution, we encode features obtained in previous layer again using a ranking machine. Compared with VideoDarwin, HiVideoDarwin captures the global and high-level video representation and is robust to large appearance changes. Compared with RNN, HiVideoDarwin can also abstract semantic information in a hierarchical way and is fast to compute and easy to interpret. We evaluate the proposed method on two datasets, namely MPII Cooking and Chalearn. Experimental results show that HiVideoDarwin has distinct advantages over the state-of-the-art models. Additional sensitivity analysis reveals that the overall results are hardly affected by parameter changes.
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/14069
专题	智能感知与计算研究中心
作者单位	1.Center for Research on Intelligent Perception and Computing (CRIPAC) 2.National Laboratory of Pattern Recognition 3.Institute of Automation, Chinese Academy of Sciences
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Wang Hongsong,Wang Wei,Wang Liang. Hierarchical Motion Evolution for Action Recognition[C],2015.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Hierarchical Motion （257KB）	会议论文		开放获取	CC BY-NC-SA	浏览