NIE Huirao,TAO Linmi.Inference framework for activity recognition based on multiple semantic layers[J].CAAI Transactions on Intelligent Systems,2015,10(02):178-186.[doi:10.3969/j.issn.1673-4785.201407009]





Inference framework for activity recognition based on multiple semantic layers
聂慧饶 陶霖密
清华大学 计算机科学与技术系, 北京 100084
NIE Huirao TAO Linmi
Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
activity recognitionfeature activity relationenvironment contextsemantic layermultilayer inference framework
人类行为理解是实现“人本计算”模式的基础,其本质在于获取行为的语义,即由动作特征推导人体的行为,需要跨越两者之间的语义鸿沟;为此提出了环境上下文进行隐式建模的方法,并基于此提出了语义分层的行为推理框架,该框架使用了从模糊语义到确定语义的渐近式推理。根据知识将特征合理地分为多个层次,系统则根据当前状态去提取所需要的特征,推理当前可能的候选行为集;并由该候选行为集指导处理模块,更新特征集并进行新一轮的推理,反复迭代至推理完成。 应用提出的环境建模方法和渐近推理框架可以有效地实现行为理解。 使用隐式环境方法可以提高行为理解的准确率;渐近式推理框架可以避免传统推理方法无差别地提取所有特征,从而提升了推理效率。
Human activity recognition is the core of the implementation of human-centered computing(HCC), whose nature is to acquire activities’ semanteme. The basic problem is the semantic gap between observable actions and human activities. They should be bridged by environment context based inference. In this paper, a method is proposed to model the environment context implicitly. Further, a novel semanteme multilayered activity inference framework was presented, which divided the inferring process into 2 stages. One stage used to acquire fuzzy semanteme and another one to acquire accurate semanteme. The feature set was divided into different subsets according to knowledge. The system extracts the corresponding features according to the current state and obtains the possible set of candidate activities that can instruct the system to update the current feature set. Update the features set and infer it, the process continues until the inference is completed. The modeling method and progressive inference framework proposed could handle the activity-recognition problem well. Implicitly modeling the environment context could improve the accuracy of activity recognition. The progressive framework can improve the efficiency by avoiding extracting all features indistinguishably, whose validity was proven in the data set.


[1] PANTIC M, PENTLAND A, NIJHOLT A, et al. Human computing and machine understanding of human behavior: a survey[C]//Proceedings of ACM International Conference on Multimodal Interfaces. Banff, Canada, 2006: 260-266.
[2] 石为人, 周彬, 许磊. 普适计算: 人本计算[J]. 计算机应用, 2005, 25(7) : 1479-1484.SHI Weiren, ZHOU Bin, XU Lei. Pervasive computing: human-centered computing[J]. Computer Applications, 2005, 25(7) : 1479-1484.
[3] 陶霖密, 杨卓宁, 王国建. 行为理解的认知方法[J]. 中国图象图形学报, 2014, 19(2) : 167-174.TAO Linmi, YANG Zhuoning, WANG Guojian. Cognitive reasoning method for behavior understanding[J]. Computer Applications, 2014, 19(2) : 167-174.
[4] SHARIAT S, PAVLOVIC V. A new adaptive segmental matching measure for human activity recognition[C]//Proceedings of IEEE International Conference on Computer Vision. Sydney, 2013: 3583-3590.
[5] BOUCHARD B, GIROUX S, BOUZOUANE A. A smart home agent for plan recognition of cognitively-impaired patients[J]. Journal of Computers, 2006, 1(5) : 53-62.
[6] CHEN L, NUGENT C D, MULVENNA M, et al. A logical framework for behavior reasoning and assistance in a smart home[J]. International Journal of Assistive Robotics and Mechatronics, 2008, 9(4) : 20-34.
[7] THOMSON G, TERZIS S, NIXON P. Situation determination with reusable situation specifications[C]//Proceedings of IEEE International Conference on Pervasive Computing and Communications Workshops. Pisa, Italy, 2006: 620-623.
[8] ISHIMARU S, UEMA Y, KUNZE K, et al. Smarter eyewear: using commercial EOG glasses for activity recognition[C]//Proceedings of ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct Publication.[S.l], 2014: 239-242.
[9] HUNH T, SCHIELE B. Unsupervised discovery of structure in activity data using multiple eigenspaces[C]//Proceedings of Second International Workshop on Location-and Context-Awareness. Dublin, Ireland, 2006: 151-167.
[10] LIAO L, FOX D, KAUTZ H. Extracting places and activities from GPS traces using hierarchical conditional random fields[J]. The International Journal of Robotics Research, 2007, 26(1) : 119-134.
[11] WARD J A, LUKOWICZ P, TROSTER G, et al. Activity recognition of assembly tasks using body-worn microphones and accelerometers[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(10) : 1553-1567.
[12] LIU C D, CHUNG Y N, CHUNG P C. An interaction-embedded HMM framework for human behavior understanding: with nursing environments as examples[J]. IEEE Transactions on Information Technology in Biomedicine, 2010, 14(5) : 1236-1246.
[13] SINGLA G, COOK D J, SCHMITTER-EDGECOMBE M. Recognizing independent and joint activities among multiple residents in smart environments[J]. Journal of Ambient Intelligence and Humanized Computing, 2010, 1(1) : 57-63.
[14] WANG S, PENTNEY W, POPESCU A M, et al. Common sense based joint training of human activity recognizers[C]//Proceedings of IJCAI. Hyderabad, India, 2007: 2237-2242.
[15] SMINCHISESCU C, KANAUJIA A, METAXAS D. Conditional models for contextual human motion recognition[J]. Computer Vision and Image Understanding, 2006, 104(2): 210-220.
[16] LEE S W, MASE K. Activity and location recognition using wearable sensors[J]. IEEE Pervasive Computing, 2002, 1(3) : 24-32.
[17] WANG G, JIANG J, SHI M. A context model for collaborative environment[C]//Proceedings of IEEE International Conference on Computer Supported Cooperative Work in Design. Nanjing, China, 2006: 1-6.
[18] LI M. Ontology-based Context information modeling for smart space[C]//Proceedings of IEEE International Conference on Cognitive Informatics and Cognitive Computing. Banff, Canada, 2011: 278-283.
[19] MOESLUND T B, HILTON A, KRüGER V. A survey of advances in vision-based human motion capture and analysis[J]. Computer Vision and Image Understanding, 2006, 104(2) : 90-126.
[20] AGGARWAL J K, PARK S. Human motion: modeling and recognition of actions and interactions[C]//Proceedings of IEEE International Symposium on 3D Data Processing, Visualization and Transmission. Thessaloniki, Greece, 2004. 640-647.
[21] GONZàLEZ J, VARONA J, ROCA F X, et al. aSpaces: Action spaces for recognition and synthesis of human actions[C]//Proceedings of Articulated Motion and Deformable Objects. Palma de Mallorca, Spain, 2002: 189-200.
[22] SUN L, DI H, TAO L, et al. A robust approach for person localization in multi-camera environment[C]//Proceedings of IEEE International Conference on Pattern Recognition. Istanbul, Turkey, 2010: 4036-4039.
[23] LUO Y, WU T D, HWANG J N. Object-based analysis and interpretation of human motion in sports video sequences by dynamic Bayesian networks[J]. Computer Vision and Image Understanding, 2003, 92(2) : 196-216.


 TIAN Guo-hui,JI Yan-qing,LI Xiao-lei.Human behaviors understanding based on scene knowledge in home intelligent space[J].CAAI Transactions on Intelligent Systems,2010,5(02):57.


更新日期/Last Update: 2015-06-15