Abstract
In activity recognition, traditionally, features are chosen heuristically, based on explicit domain knowledge. Typical features are statistical measures, like mean, standard deviation, etc., which are tailored to the application at hand and might not fit in other cases. However, Feature Learning techniques have recently gained attention for building approaches that generalize over different application domains. More conventional approaches, like Principal Component Analysis, and newer ones, like Deep Belief Networks, have been studied so far and yielded significantly better results than traditional techniques. In this paper we study the potential of Shift-invariant Sparse Coding (SISC) as an additional Feature Learning technique for activity recognition. We evaluate the performance on several publicly available activity recognition data sets and show that classification based on features learned by SISC outperforms other previously presented Feature Learning techniques.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Eggert, J., Wersing, H., Korner, E.: Transformation-invariant representation and NMF. In: Proc. of the 2004 IEEE Int. Joint Conf. on Neural Networks, vol. 4, pp. 2535–2539. IEEE (2004)
Huynh, T., Schiele, B.: Analyzing features for activity recognition. In: Proc. of the 2005 Joint Conf. on Smart Objects and Ambient Intelligence (2005)
Huynh, T., Fritz, M., Schiele, B.: Discovery of activity patterns using topic models. In: Proc. of the 10th Int. Conf. on Ubiquitous Computing, pp. 10–19. ACM Press, New York (2008)
Kim, T., Shakhnarovich, G., Urtasun, R.: Sparse Coding for Learning Interpretable Spatio-Temporal Primitives. In: Proc. Neural Inf. Process. Syst (NIPS), vol. 22 (December 2010)
Lee, D.D., Seung, H.S.: Learning the parts of objects by non-negative matrix factorization. Nature 401(6755), 788–791 (1999)
Pham, C., Olivier, P.: Slice&Dice: Recognizing food preparation activities using embedded accelerometers. In: Tscheligi, M., de Ruyter, B., Markopoulus, P., Wichert, R., Mirlacher, T., Meschterjakov, A., Reitberger, W. (eds.) AmI 2009. LNCS, vol. 5859, pp. 34–43. Springer, Heidelberg (2009)
Plötz, T., Hammerla, N.Y., Olivier, P.: Feature Learning for Activity Recognition in Ubiquitous Computing. In: Proc. of the Twenty-Second Int. Joint Conf. on Artificial Intelligence, pp. 1729–1734 (2011)
Roggen, D., et al.: Collecting complex activity datasets in highly rich networked sensor environments. In: 2010 Seventh Int. Conf. on Networked Sensing Systems (INSS), pp. 233–240 (June 2010)
Smith, E., Lewicki, M.S.: Efficient coding of time-relative structure using spikes. Neural Computation 17(1), 19–45 (2005)
Vollmer, C., Eggert, J.P., Groß, H.-M.: Generating Motion Trajectories by Sparse Activation of Learned Motion Primitives. In: Villa, A.E.P., Duch, W., Érdi, P., Masulli, F., Palm, G. (eds.) ICANN 2012, Part I. LNCS, vol. 7552, pp. 637–644. Springer, Heidelberg (2012)
Zappi, P., Lombriser, C., Stiefmeier, T., Farella, E., Roggen, D., Benini, L., Tröster, G.: Activity recognition from on-body sensors: Accuracy-power trade-off by dynamic sensor selection. In: Verdone, R. (ed.) EWSN 2008. LNCS, vol. 4913, pp. 17–33. Springer, Heidelberg (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Vollmer, C., Gross, HM., Eggert, J.P. (2013). Learning Features for Activity Recognition with Shift-Invariant Sparse Coding. In: Mladenov, V., Koprinkova-Hristova, P., Palm, G., Villa, A.E.P., Appollini, B., Kasabov, N. (eds) Artificial Neural Networks and Machine Learning – ICANN 2013. ICANN 2013. Lecture Notes in Computer Science, vol 8131. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-40728-4_46
Download citation
DOI: https://doi.org/10.1007/978-3-642-40728-4_46
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-40727-7
Online ISBN: 978-3-642-40728-4
eBook Packages: Computer ScienceComputer Science (R0)