Abstract
Low quality images taken by surveillance cameras pose a great challenge to human action recognition algorithms. This is because they are usually noisy, of low resolution and of low frame rate. In this paper we propose an action recognition algorithm to overcome the above challenges. We use optic flow to construct motion descriptors and apply a SVM to classify them. Having powerful discriminative features, we significantly reduce the size of the feature set required. This algorithm can be applied to videos with low frame rate without scarifying efficiency or accuracy, and is robust to scale and view point changes. To evaluate our method, we used a database consisting of walking, running, jogging, hand clapping, hand waving and boxing actions. This grayscale database has images of low resolution and poor quality. This image database resembles images taken by surveillance cameras. The proposed method outperforms competing algorithms evaluated on the same database.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ahammad, P., Yeo, C., Sastry, S.S., Ramchandran, K.: Compressed Domain Real-time Action Recognition. In: MMSP. Proceedings of 8th IEEE Workshop on Multimedia Signal Processing, pp. 33–36. IEEE Computer Society Press, Los Alamitos (2006)
Black, M.J., Anandan, P.: The Robust Estimation of Multiple Motions: Parametric and Piecewise-Smooth Flow Fields. Computer Vision and Image Understanding (CVIU) 63, 75–104 (1996)
Dollar, P., Rabaud, V., Cottrel, G., Belongie, S.: Behavior Recognition Via Spatio-Temporal Features. In: Proceedings 2nd Joint IEEE International Workshop on VS-PETS, Beijing, pp. 65–72. IEEE Computer Society Press, Los Alamitos (2005)
Efros, A.A., Berg, A.C., Mori, G., Malik, J.: Recognizing Action at a Distance. In: ICCV 2003. IEEE International Conference on Computer Vision, Nice, France, pp. 726–733. IEEE Computer Society Press, Los Alamitos (2003)
Gavrila, D.M.: The Analysis of Human Motion and its Application for Visual Surveillance. In: Proceedings of the Second IEEE Workshop on Visual Surveillance, p. 3. IEEE Computer Society Press, Los Alamitos (1999)
Gavrila, D.M.: The Visual Analysis of Human Movement: A Survey. Computer Vision and Image Understanding (CVIU) 23, 82–98 (1999)
Harris, C., Stephens, M.: A combined Corner and Edge Detector. In: Proceedings of the 4th Alvey Vision Conference, Manchester, UK, pp. 147–151 (1988)
Horn, B.K., Schunck, B.G.: Determining Optic Flow. Artificial Intelligence 17, 185–203 (1981)
Chih-Jen, L., Chin-chung, C.: LIBSVM–A Library for Support Vector Machines (2007)
Niebles, J.C., Wang, H., Fei-fei, L.: Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words. In: BMVC (2006)
Niu, W., Long, J., Han, D., Wang, Y.-F.: Human Activity Detection and Recognition for Video Surveillance. In: Proceedings of the IEEE Multimedia and Expo Conference, pp. 719–722. IEEE Computer Society Press, Los Alamitos (2004)
Robertson, N., Reid, I., Brady, M.: Behavior Recognition and Explanation for Video Survillance. In: ICDP 2006. IEE Conf. Imaging for Crime Detection and Prevention (2006)
Schuldt, C., Laptev, I., Caputo, B.: Recognizing Human Actions: A Local SVM Approach. In: ICPR 2004, pp. 32–36 (2004)
Swain, M.J., Ballard, D.H.: Color Indexing. International Journal of Computer Vision 7, 11–32 (1991)
Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)
Wang, L., Ning, H., Hu., W.: Fusion of Static and Dynamic Body Biometrics for Gait Recognition. In: International Conference on Computer Vision, Nice, France, pp. 1449–1454 (2003)
Yilmaz, A., Shah, M.: Actions As Objects: A Novel Action Representation. In: IEEE CVPR, San Diego, IEEE Computer Society Press, Los Alamitos (2005)
Zhu, G., Changsheng, X., Gao, W., Huang, Q.: Action Recognition in Broadcast Tennis Video Using Optical Flow and SVM. In: Proceedings of HCI/ECCV (2006)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Danafar, S., Gheissari, N. (2007). Action Recognition for Surveillance Applications Using Optic Flow and SVM. In: Yagi, Y., Kang, S.B., Kweon, I.S., Zha, H. (eds) Computer Vision – ACCV 2007. ACCV 2007. Lecture Notes in Computer Science, vol 4844. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-76390-1_45
Download citation
DOI: https://doi.org/10.1007/978-3-540-76390-1_45
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-76389-5
Online ISBN: 978-3-540-76390-1
eBook Packages: Computer ScienceComputer Science (R0)