[1]庄伟源,成运,林贤明,等.关键肢体角度直方图的行为识别[J].智能系统学报,2015,10(1):20-26.[doi:10.3969/j.issn.1673-4785.201410039]
　ZHUANG Weiyuan,CHENG Yun,LIN Xianming,et al.Action recognition based on the angle histogram of key parts[J].CAAI Transactions on Intelligent Systems,2015,10(1):20-26.[doi:10.3969/j.issn.1673-4785.201410039]
点击复制

关键肢体角度直方图的行为识别

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 10 期数: 2015年第1期 页码: 20-26 栏目: 学术论文—智能系统 出版日期: 2015-03-25
Title:
Action recognition based on the angle histogram of key parts
作者:
庄伟源1,3, 成运2, 林贤明1,3, 苏松志1,3, 曹冬林1,3, 李绍滋1,3
1. 厦门大学 信息科学与技术学院, 福建 厦门 361005;
2. 湖南人文科技学院 通信与控制工程系, 湖北 娄底 417000;
3. 福建省仿脑智能系统重点实验室, 福建 厦门 361005
Author(s):
ZHUANG Weiyuan1,3, CHENG Yun2, LIN Xianming1,3, SU Songzhi1,3, CAO Donglin1,3, LI Shaozi1,3
1. School of Information Science and Technology, Xiamen University, Xiamen 361005, China;
2. Department of Communication and Control Engineering, Hunan University of Humanities, Science and Technology, Loudi 417000, China;
3. Fujian Key Laboratory of the Brain-Like Intelligent Systems, Xiamen 361005, China
关键词:
角度特征; 动作识别; 关键肢体; 角度直方图; 姿态表示; 行为分析; 动作特征
Keywords:
angle feature; action recognition; key parts; angle histogram; pose representation; action analyze; action feature
分类号:
TP391.4
DOI:
10.3969/j.issn.1673-4785.201410039
文献标志码:
A
摘要:
当前的姿态表示的行为识别方法通常对姿态的准确性做了很强的假设,而当姿态分析不精确时,这些现有方法的识别效果不佳。提出了一种低维的、鲁棒的基于关键肢体角度直方图的人体姿态特征描述子,用于将整个动作视频映射成一个特征向量。同时,还在特征向量中引入共生模型,用以表示肢体间的关联性。最后,设计了分层的SVM分类器,第1层主要用于选择高判别力的肢体作为关键肢体,第2层则利用关键肢体的角度直方图并作为特征向量,进行行为识别。实验结果表明,基于关键肢体角度直方图的动作特征具有较好的判别能力,能更好地区分相似动作,并最终取得了更好的识别效果。
Abstract:
The current pose-based methods usually make a strong assumption for the accuracy of pose, but when the pose analysis is not precise, these methods cannot achieve satisfying results of recognition. Therefore, this paper proposed a low-dimensional and robust descriptor on the gesture feature of the human body based on the angle histogram of key limbs, which is used to map the entire action video into an feature vector. A co-occurrence model is introduced into the feature vector for expressing the relationship among limbs. Finally, a two-layer support vector machine (SVM) classifier is designed. The first layer is used to select highly discriminative limbs as key limbs and the second layer takes angle histogram of key limbs as the feature vector for action recognition. Experiment results demonstrated that the action feature based on angle histogram of key limbs has excellent judgment ability, may properly distinguish similar actions and achieve better recognition effect.
参考文献/References:
[1] YANG Weilong, WANG Yang, MORI G. Recognizing human actions from still images with latent poses[C]//IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, USA, 2010: 2030-2037.
[2] LAPTEV I. On space-time interest points[J]. International Journal of Computer Vision, 2005, 64(2/3): 107-123.
[3] WANG H, KLASER A, SCHMID C, et al. Action recognition by dense trajectories[C]//IEEE Conference on Computer Vision and Pattern Recognition. Colorado, USA, 2011: 3169-3176.
[4] KLASER A, MARSZALEK M, SCHMID C. A spatio-temporal descriptor based on 3d-gradients[C]//British Machine Vision Conference. Leeds, UK, 2008: 275-285.
[5] SADANAND S, CORSO J. Action bank: a high-level representation of activity in video[C]//IEEE Conference on Computer Vision and Pattern Recognition.[s.l.], 2012: 1234-1241.
[6] BAYSAL S, DUYGULU P. A line based pose representation for human action recognition[J]. Signal Processing: Image Communication, 2013, 28(5): 458-471.
[7] LI Wang, LI Cheng. Human action recognition from boosted pose estimation[C]//International Conference on Digital Image Computing: Techniques and Applications. Sydney, AU, 2010: 308-313.
[8] 徐光祐,曹媛媛. 动作识别和行为理解综述[J]. 中国图像图形学报, 2009,14(2): 189-195.XU Guangyou, CAO Yuanyuan. Action recognition and activity understanding: a review[J]. Journal of Image and Graphics, 2009, 14(2): 189-195.
[9] BOURDEV L, MALIK J. Poselets: body part detectors training using 3-D human pose annotations[C]//IEEE International Conference on Computer Vision.[s.l.], 2009: 1365-1372.
[10] FELZENSZWALB P, MCALLESTER D, RAMANAN D. A discriminatively trained, multi scale, deformable part model[C]//IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA, 2008: 1-8.
[11] YANG Y, RAMANAN D. Articulated pose estimation with flexible mixtures-of-parts[C]//IEEE Conference on Computer Vision and Pattern Recognition. Colorado, USA, 2011: 1385-1392.
[12] YANG Y, RAMANAN D. Articulated human detection with flexible mixtures of parts[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(12): 2878-2890.
[13] WANG Jiang, LIU Zicheng, WU Ying. Ming acionlet ensemble for action recognition with depth cameras[C]//IEEE Conference on Computer Vision and Pattern Recognition.[s.l.], USA, 2012: 1290-1297.
[14] 雷庆,李绍滋. 动作识别中局部时空特征的运动表示方法研究[J]. 计算机工程与应用, 2010, 46(34): 7-10.LEI Qing, LI Shaozi. Research on local spatio-temporal features for action recognition[J]. Computer Engineering and Applications, 2010, 46(34): 7-10.
[15] EPSHTEIN B, ULLMAN S. Semantic hierarchies for recognizing objects and parts[C]//IEEE Conference on Computer Vision and Pattern Recognition., 2007: 1-8.
[16] FELZENSZWALB P, HUTTENLOCHER D. Pictorial structures for object recognition[J]. International Journal of Computer Vision, 2005, 61(1): 55-79.
[17] FELZENSZWALB P, GIRSHICK R B, MCALLESTER D, et al. Object detection with discriminatively trained part based models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9): 1627-1645.
[18] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//IEEE Conference on Computer Vision and Pattern Recognition., 2005: 886-893.
[19] 曲永宇,刘清,郭建明. 基于HOG和颜色特征的行人检测[J]. 武汉理工大学学报, 2011, 33(4):134-141.QU Yongyu, LIU Qing, GUO Jianming. HOG and color based pedestrian detection[J]. Journal of Wuhan University of Technology, 2011, 33(4): 134-141.
[20] LAPTEV I, CAPUTO B, SCHULDT Christian. Local velocity-adapted motion events for spatio-temporal recognition[J]. Computer Vision and Image Understanding, 2007, 108: 207-229.
[21] BURGOS-ARTIZZU X P, HALL D, PIETRO P, et al. Merging pose estimates across space and time[C]//British Machine Vision Conference. Bristol, UK, 2013: 58-69.
[22] 王千,王成,冯振元. K-means聚类算法研究综述[J]. 电子设计工程, 2012, 20(7): 21-24.WANG Qian, WANG Cheng, FENG Zhenyuan. Review of K-means cluster algorithm[J]. Electronic Design Engineering, 2012, 20(7): 21-24.
[23] LAPTEV I, MARSZALEK M, SCHMID C, el al. Learning realistic human actions from movies[C]//IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA, 2008: 1-8.
[24] BREGONZIO M, GONG S, XIANG T. Recognizing action as clouds of space-time interest points[C]//IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA, 2009: 1948-1955.
[25] LIU J, SHAH M. Learning human actions via information maximization[C]//IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA, 2008: 1-8.
[26] WU X, XU D, DUAN L,et al. Action recognition using context and appearance distribution features[C]//IEEE Conference on Computer Vision and Pattern Recognition. Colorado, USA, 2011: 489-496.
[27] GILBERT A, ILLINGWORTH J, BOWDEN R. Fast realistic multi-action recognition using mined dense spatio-temporal features[C]//IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA, 2009: 925-931.
[28] 凌志刚,赵春晖,梁彦. 基于视觉的人行为理解综述[J]. 计算机应用研究, 2008, 25(9): 2570-2578.LING Zhigang, ZHAO Chunhui, LIANG Yan. Survey on vision-based human action understanding[J]. Application Research of Computers, 2008, 25(9): 2570-2578.
相似文献/References:
[1]姬晓飞,王昌汇,王扬扬.分层结构的双人交互行为识别方法[J].智能系统学报,2015,10(6):893.[doi:10.11992/tis.201505006]
　JI Xiaofei,WANG Changhui,WANG Yangyang.Human interaction behavior-recognition method based on hierarchical structure[J].CAAI Transactions on Intelligent Systems,2015,10():893.[doi:10.11992/tis.201505006]
[2]王策,姬晓飞,李一波.一种简便的视角无关动作识别方法[J].智能系统学报,2014,9(5):577.[doi:10.3969/j.issn.1673-4785.201307057]
　WANG Ce,JI Xiaofei,LI Yibo.Study on a simple view-invariant action recognition method[J].CAAI Transactions on Intelligent Systems,2014,9():577.[doi:10.3969/j.issn.1673-4785.201307057]
[3]莫凌飞,蒋红亮,李煊鹏.基于深度学习的视频预测研究综述[J].智能系统学报,2018,13(1):85.[doi:10.11992/tis.201707032]
　MO Lingfei,JIANG Hongliang,LI Xuanpeng.Review of deep learning-based video prediction[J].CAAI Transactions on Intelligent Systems,2018,13():85.[doi:10.11992/tis.201707032]
[4]钟秋波,郑彩明,朴松昊.时空域融合的骨架动作识别与交互研究[J].智能系统学报,2020,15(3):601.[doi:10.11992/tis.202006029]
　ZHONG Qiubo,ZHENG Caiming,PIAO Songhao.Research on skeleton-based action recognition with spatiotemporal fusion and human–robot interaction[J].CAAI Transactions on Intelligent Systems,2020,15():601.[doi:10.11992/tis.202006029]

备注/Memo

收稿日期:2014-10-24;改回日期:。
基金项目:国家自然科学基金资助项目(61202143);福建省自然科学基金资助项目(2013J05100,2010J01345,2011J01367);厦门市科技重点项目资助项目(3502Z20123017).
作者简介:庄伟源,男,1990年生,硕士研究生,主要研究方向为人体行为识别、计算机视觉、深度学习;林贤明,男,1980年生,助理教授,博士,主要研究方向为人体行为识别、移动视觉搜索、计算机视觉、模式识别;李绍滋,男,1963年生,教授,博士生导师,博士,福建省人工智能学会副理事长兼秘书长,主要研究方向为运动目标检测与识别、自然语言处理与多媒体信息检索等。发表学术论文160余篇,其中被SCI检索16篇、被EI检索142篇。
通讯作者:林贤明.E-mail:linxm@xmu.edu.cn.

更新日期/Last Update: 2015-06-16
Copyright © 《 智能系统学报》 编辑部
地址：（150001）黑龙江省哈尔滨市南岗区南通大街145-1号楼 电话：0451- 82534001、82518134　邮箱：tis@vip.sina.com