International Journal of Information Technology & Computer Science ( IJITCS )
A novel approach is proposed in this work for human activity recognition from depth video sensor data using features of human body and two-level Hidden Markov Models. From depth video, different body parts of human activities are segmented first by means of random forests. Then, robust features are obtained using labeled body parts and body joint information which includes motion information as well. Traditionally, a dictionary of all trained HMMs for all the activities are built and the features are applied on the HMMs to generate the maximum likelihood to obtain the proper activity which is time consuming if the activities are many. Hence, two-level activity recognition is proposed in this work. Initially, all activities are distributed in some groups. In first level, activity group is determined by means of applying the features on activity class HMMs. Finally, the features are applied again on the trained activity HMMs of the group obtained from first level classification. The proposed approach shows better performance than traditional approaches.
: Body Joint, Depth Information, Hidden Markov Models.
- M. Z. Uddin, D. H. Kim, J. T. Kim, and T.-S. Kim, “An Indoor Human Activity Recognition System for Smart Home Using Local Binary Pattern Features with Hidden Markov Models,” Indoor and Built Environment, vol. 22, pp. 289-298, 2013.
- A. Jalal, M. Z. Uddin, J. J. Lee, and T.-S. Kim, “Recognition of Human Home Activities via Depth Silhouettes and R Transformation for Smart Home,” Indoor and building Environment, vol. 21(1), pp. 184- 190, 2012.
- M. Z. Uddin, J.J. Lee, and T.-S Kim, “Independent shape component-based human activity recognition via Hidden Markov Model,” Journal of Applied Intelligence, pp. 193-206, 2010.
- P. Simari, D. Nowrouzezahrai, E. Kalogerakis, and K. Singh, “Multi-objective shape segmentation and labeling,” Eurographics Symposium on Geometry Processing, vol. 28, pp. 1415-1425, 2009.
- V. Ferrari, M.-M. Jimenez, and A. Zisserman, “2D Human Pose Estimation in TV Shows,” Visual Motion Analysis, LNCS 5604, pp. 128-147, 2009.
- H. Lu, K. N. Plataniotis, and A. N. Venetsanopoulos, “A Full-Body Layered Deformable Model for Automatic Model-Based Gait Recognition,” EURASIP Journal on Advances in Signal Processing, pp. 1-13, 2008.
- J. Wright, and G. Hua, “Implicit Elastic Matching with Random Projections for Pose-Variant face recognition,” IEEE conf. on Computer Vision and Pattern Recognition, pp. 1502-1509, 2009.
- A. Bosch, A. Zisserman, and X. Munoz, “Image classification using random forests and ferns,” IEEE Int. Conf. on Computer Vision, pp. 1-8, 2007.
- Microsoft Corporation: ‘Kinect for Xbox 360-Xbox.com’, http://www.xbox.com/en-GB/kinect/, Accessed September 4, 2014.
- V. Lepetit, and P. Fua, “Keypoint recognition using randomized trees,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 28, pp. 1465-1479, 2006.
- T. Kanungu, D. M. Mount, N. Netanyahu, C. Piatko, R. Silverman, and A. Y. Wu, “The analysis of a simple k-means clustering algorithm”, in Proceedings of 16th ACM Symposium On Computational Geometry, pp. 101-109, 2000.
- Y. Linde, A. Buzo, and R. Gray, “An algorithm for vector quantizer design”, IEEE Transaction on Communications, vol. 28, no. 1, pp. 84-94, 1980.
- E. Baum, T. Petrie, G. Soules, and N. Weiss, “A maximization technique occurring in the statistical analysis of probabilistic functions of markov chains”, Annals of Mathematical Statistics, vol. 41, pp. 164-171, 1970.
- E. Baum and J. Eagon, “An inequality with applications to statistical estimation for probabilistic functions of markov processes and to a model for ecology”, American Mathematical Society Bulletin, vol. 73, pp. 360-363, 1967.
- R. Lawrence and A. Rabiner, “Tutorial on hidden markov models and selected applications in speech recognition”, in Proceedings of the IEEE, vol. 77, No. 2, pp. 257-286, 1989.