第3期 钟秋波，等：时空域融合的骨架动作识别与交互研究 ·607· 别，可以适应复杂的背景条件，体验者的交互效 havior detection of joint weighted reconstruction trajectory 率更高。 and histogram entropy[J].CAAI transactions on intelligent systems,.2018,13(6):1015-1026. 参考文献： [10]吴云鹏，赵晨阳，时增林，等.基于流密度的多重交互集 [1]SIMONYAN K,ZISSERMAN A.Two-stream convolu- 体行为识别算法[】.计算机学报，2017,40(11)： tional networks for action recognition in videos[C]//Pro- 2519-2532. ceedings of the 27th International Conference on Neural WU Yunpeng,ZHAO Chenyang,SHI Zenglin,et al.A Information Processing Systems.Montreal,Canada,2014: flow density based algorithm for detecting coherent mo- 568-576. tion with multiple interaction[.Chinese journal of com- [2]BAGAUTDINOV T,ALAHI A,FLEURET F,et al.So- puters,2017,40(11):2519-2532. cial scene understanding:end-to-end multi-person action [11]陈婷婷，阮秋琦，安高云.视频中人体行为的慢特征提 localization and collective activity recognition[C]//Pro- 取算法.智能系统学报，2015,10(3)：381-386 ceedings of the IEEE Conference on Computer Vision and CHEN Tingting,RUAN Qiuqi,AN Gaoyun.Slow fea- Pattern Recognition.Honolulu,USA,2017:3425-3434. ture extraction algorithm of human actions in video[J]. [3]WANG Heng,SCHMID C.Action recognition with im- CAAI transactions on intelligent systems,2015,10(3): proved trajectories[C]//Proceedings of the IEEE Interna- 381-386. tional Conference on Computer Vision.Sydney,Australia, [12]丁重阳，刘凯，李光，等.基于时空权重姿态运动特征的 2013:3551-3558 人体骨架行为识别研究U.计算机学报，2020,43(1)： [4]CAO Zhe.SIMON T,WEI S E,et al.Realtime multi-per- 29-40. son 2D pose estimation using part affinity fields[C]//Pro- DING Chongyang,LIU Kai,LI Guang,et al.Spatio-tem- ceedings of the IEEE Conference on Computer Vision and poral weighted posture motion features for human skelet- Pattern Recognition.Honolulu,USA,2017:1302-1310. on action recognition research[J].Chinese journal of com- [5]CHEN Yilun,WANG Zhicheng,PENG Yuxiang,et al. puters,2020,43(1:29-40. Cascaded pyramid network for multi-person pose estima- [13]莫宏伟，汪海波.基于Faster R-CNN的人体行为检测研 tion[C]//Proceedings of the IEEE/CVF Conference on 究U.智能系统学报，2018,13(6)：967-973 Computer Vision and Pattern Recognition.Salt Lake City, MO Hongwei,WANG Haibo.Research on human beha- USA,2018:7103-7112. vior detection based on Faster R-CNN[J].CAAl transac- [6]龚冬颖，黄敏，张洪博，等.RGBD人体行为识别中的自 tions on intelligent systems,2018,13(6):967-973. 适应特征选择方法仞.智能系统学报，2017,12(1)：1-7. [14]姬晓飞，谢旋，任艳.深度学习的双人交互行为识别与 GONG Dongying,HUANG Min,ZHANG Hongbo,et al. 预测算法研究.智能系统学报，DO:10.11992is. Adaptive feature selection method for action recognition of 201812029. human body in RGBD data[J].CAAI transactions on intel- JI Xiaofei,XIE Xuan,Ren Yan.Human interaction recog- ligent systems,2017,12(1):1-7. nition and prediction algorithm based on Deep Learning [7]姬晓飞，王昌汇，王扬扬.分层结构的双人交互行为识别 [J].CAAI transactions on intelligent systems,DOI: 方法[).智能系统学报，2015,10(6：893-900 10.11992is.201812029 JI Xiaofei,WANG Changhui,WANG Yangyang.Human [15]谢昭，周义，吴克伟，等.基于时空关注度LSTM的行为 interaction behavior-recognition method based on hierarch- 识别[J/OL].计算机学报：(2019-12-17)http:/kns ical structure[J].CAAI transactions on intelligent systems, cnki.net/kcms/detail/11.1826.TP.20191227.1658.002.html. 2015,10(6):893-900. XIE Zhao,ZHOU Yi,WU Kewei,et al.Activity recogni- [8]庄伟源，成运，林贤明，等.关键肢体角度直方图的行为 tion based on spatial-temporal attention LSTM[J/OL] 识别).智能系统学报，2015,10(1)：20-26. Chinese journal of computers:(2019-12-17)http://kns. ZHUANG Weiyuan,CHENG Yun,LIN Xianming,et al. cnki.net/kems/detail/11.1826.TP.20191227.1658.002.html. Action recognition based on the angle histogram of key [16]王传旭，胡小悦，孟唯佳，等.基于多流架构与长短时记 parts[J].CAAI transactions on intelligent systems,2015, 忆网络的组群行为识别方法研究).电子学报，2020， 10(1):20-26. 48(4:800-807 [9]徐志通，骆炎民，柳培忠.联合加权重构轨迹与直方图嫡 WANG Chuanxu,HU Xiaoyue,MENG Weijia,et al.Re- 的异常行为检测[J】.智能系统学报，2018,13(6)： search on group behavior recognition method based on 1015-1026 multi-stream architecture and long short-term memory XU Zhitong,LUO Yanmin,LIU Peizhong.Abnormal be- network[J].Acta electronica sinica,2020,48(4):800-807.别，可以适应复杂的背景条件，体验者的交互效 率更高。 参考文献： SIMONYAN K, ZISSERMAN A. Two-stream convolu￾tional networks for action recognition in videos[C]//Pro￾ceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada, 2014: 568−576. [1] BAGAUTDINOV T, ALAHI A, FLEURET F, et al. So￾cial scene understanding: end-to-end multi-person action localization and collective activity recognition[C]//Pro￾ceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 3425−3434. [2] WANG Heng, SCHMID C. Action recognition with im￾proved trajectories[C]//Proceedings of the IEEE Interna￾tional Conference on Computer Vision. Sydney, Australia, 2013: 3551−3558. [3] CAO Zhe, SIMON T, WEI S E, et al. Realtime multi-per￾son 2D pose estimation using part affinity fields[C]//Pro￾ceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 1302−1310. [4] CHEN Yilun, WANG Zhicheng, PENG Yuxiang, et al. Cascaded pyramid network for multi-person pose estima￾tion[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018: 7103−7112. [5] 龚冬颖, 黄敏, 张洪博, 等. RGBD 人体行为识别中的自 适应特征选择方法 [J]. 智能系统学报, 2017, 12(1): 1–7. GONG Dongying, HUANG Min, ZHANG Hongbo, et al. Adaptive feature selection method for action recognition of human body in RGBD data[J]. CAAI transactions on intel￾ligent systems, 2017, 12(1): 1–7. [6] 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): 893–900. JI Xiaofei, WANG Changhui, WANG Yangyang. Human interaction behavior-recognition method based on hierarch￾ical structure[J]. CAAI transactions on intelligent systems, 2015, 10(6): 893–900. [7] 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. 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. [8] 徐志通, 骆炎民, 柳培忠. 联合加权重构轨迹与直方图熵 的异常行为检测 [J]. 智能系统学报, 2018, 13(6): 1015–1026. XU Zhitong, LUO Yanmin, LIU Peizhong. Abnormal be- [9] havior detection of joint weighted reconstruction trajectory and histogram entropy[J]. CAAI transactions on intelligent systems, 2018, 13(6): 1015–1026. 吴云鹏, 赵晨阳, 时增林, 等. 基于流密度的多重交互集 体行为识别算法 [J]. 计算机学报, 2017, 40(11): 2519–2532. WU Yunpeng, ZHAO Chenyang, SHI Zenglin, et al. A flow density based algorithm for detecting coherent mo￾tion with multiple interaction[J]. Chinese journal of com￾puters, 2017, 40(11): 2519–2532. [10] 陈婷婷, 阮秋琦, 安高云. 视频中人体行为的慢特征提 取算法 [J]. 智能系统学报, 2015, 10(3): 381–386. CHEN Tingting, RUAN Qiuqi, AN Gaoyun. Slow fea￾ture extraction algorithm of human actions in video[J]. CAAI transactions on intelligent systems, 2015, 10(3): 381–386. [11] 丁重阳, 刘凯, 李光, 等. 基于时空权重姿态运动特征的 人体骨架行为识别研究 [J]. 计算机学报, 2020, 43(1): 29–40. DING Chongyang, LIU Kai, LI Guang, et al. Spatio-tem￾poral weighted posture motion features for human skelet￾on action recognition research[J]. Chinese journal of com￾puters, 2020, 43(1): 29–40. [12] 莫宏伟, 汪海波. 基于 Faster R-CNN 的人体行为检测研 究 [J]. 智能系统学报, 2018, 13(6): 967–973. MO Hongwei, WANG Haibo. Research on human beha￾vior detection based on Faster R-CNN[J]. CAAI transac￾tions on intelligent systems, 2018, 13(6): 967–973. [13] 姬晓飞, 谢旋, 任艳. 深度学习的双人交互行为识别与 预测算法研究 [J]. 智能系统学报, DOI: 10.11992/tis. 201812029. JI Xiaofei, XIE Xuan, Ren Yan. Human interaction recog￾nition and prediction algorithm based on Deep Learning [J]. CAAI transactions on intelligent systems, DOI: 10.11992/tis. 201812029. [14] 谢昭, 周义, 吴克伟, 等. 基于时空关注度 LSTM 的行为 识别 [J/OL]. 计算机学报: (2019-12-17) http://kns. cnki.net/kcms/detail/11.1826.TP.20191227.1658.002.html. XIE Zhao, ZHOU Yi, WU Kewei, et al. Activity recogni￾tion based on spatial-temporal attention LSTM[J/OL] Chinese journal of computers: (2019-12-17) http://kns. cnki.net/kcms/detail/11.1826.TP.20191227.1658.002.html. [15] 王传旭, 胡小悦, 孟唯佳, 等. 基于多流架构与长短时记 忆网络的组群行为识别方法研究 [J]. 电子学报, 2020, 48(4): 800–807. WANG Chuanxu, HU Xiaoyue, MENG Weijia, et al. Re￾search on group behavior recognition method based on multi-stream architecture and long short-term memory network[J]. Acta electronica sinica, 2020, 48(4): 800–807. [16] 第 3 期 钟秋波，等：时空域融合的骨架动作识别与交互研究 ·607·