ABSTRACT
Activity Recognition is an integral component of ubiquitous computing. Recognizing an activity is a challenging task since activities can be concurrent, interleaved or ambiguous and can consist of multiple actors (which would require parallel activity recognition). This paper investigates how the discriminative nature of Conditional Random Fields (CRF) can be exploited to enhance the accuracy of recognizing activities when compared to that achieved using generative models. It aims to apply CRF to recognize complex activities, analyze the model trained by CRF and evaluate the performance of CRF against existing models using Stochastic Gradient Descent (which is suitable for online learning).
- Qing-Bin Gao, and Shi-Liang Sun. 2012. "Trajectory-Based Human Activity Recognition Using Hidden Conditional Random Fields". In Proceedings of the 2012 International Conference on Machine Learning and Cybernetics, Xian, 15--17 July, 2012. IEEE Xplore. DOI=10.1109/ICMLC.2012.6359507Google Scholar
- Eunju Kim, Sumi Helal, and Diane Cook. 2010. "Human Activity Recognition and Pattern Discovery". IEEE Pervasive Computing 9, 1 (January 2010), 48--53. DOI=10.1109/MPRV.2010.7. Google ScholarDigital Library
- Machine Learning: Generative and Discriminative Models. 2010. "Machine Learning: Generative and Discriminative Models". Machine Learning Course, http://www.cedar.buffalo.edu/~srihari/CSE574/Discriminative-Generative.pdfGoogle Scholar
- Hongqing Fang, Raghavendiran Srinivasan and Diane J. Cook, 2012, "Feature selections for human activity recognition in smart home environments". In proceedings of International Journal of Innovative Computing, Information and Control, May 2012.Google Scholar
- Sutton, Charles and Mccallum, Andrew. 2006. "Introduction to Conditional Random Fields for Relational Learning". Getoor, Lise and Taskar, Ben, eds. Introduction to Statistical Relational Learning, MIT Press.Google Scholar
- Charles Elkan. 2012. "Log Linear Models and Conditional Random Fields". University of San Diego, http://cseweb.ucsd.edu/users/elkan/250Bwinter2012/loglinear- CRFs.pdf.Google Scholar
- Douglas L. Vail, Manuela M. Veloso, and John D. Lafferty.2007. "Feature Selection in Conditional Random Fields for Activity Recognition". In Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, San Diego, CA, 2007. IROS 2007. IEEE Xplore, San Diego, CA, pp 3379--3384, DOI=10.1109/IROS.2007.4399441.Google Scholar
- Douglas L. Vail. 2008. "Conditional Random Fields for Activity Recognition". PhD Thesis, Carnegie Mellon University, United States of America, CMU-CS-08-119, April, 2008. Google ScholarDigital Library
- "Center for Advanced Studies in Adaptive Systems (CASAS)", Washington State University Smart Home Project, Aruba Dataset, 2010--2011, Tulum Dataset, 2009--2010, Two-R (Summer and Spring) Dataset 2009, Kyoto (ADL Normal and ADL Error Dataset). Available at CASAS, WSU website: http://wsucasas.wordpress.com/datasets/Google Scholar
- Naoaki Okazaki, 2007, "CRFsuite: a fast implementation of Conditional Random Fields (CRFs)", http://www.chokkan.org/software/crfsuiteGoogle Scholar
- Megha Agarwal, 2014, "MSc Thesis on Activity Recognition Using Conditional Random Fields", University Of Bristol, Bristol, United KingdomGoogle Scholar
Index Terms
- Activity recognition using conditional random field
Recommendations
Conditional random fields for activity recognition in smart environments
IHI '10: Proceedings of the 1st ACM International Health Informatics SymposiumOne of the most common functions of smart environments is to monitor and assist older adults with their activities of daily living. Activity recognition is a key component in this application. It is essentially a temporal classification problem which ...
Video-based activity level recognition for assisted living using motion features
ICDSC '15: Proceedings of the 9th International Conference on Distributed Smart CamerasActivities of daily living of the elderly is often monitored using passive sensor networks. With the reduction of camera prices, there is a growing interest of video-based approaches to provide a smart, safe and independent living environment for the ...
Conditional random fields for activity recognition
AAMAS '07: Proceedings of the 6th international joint conference on Autonomous agents and multiagent systemsActivity recognition is a key component for creating intelligent, multi-agent systems. Intrinsically, activity recognition is a temporal classification problem. In this paper, we compare two models for temporal classification: hidden Markov models (HMMs)...
Comments