Abstract
Surveillance videos of public places often consist of group activities composed from multiple co-occurring individual activities. However, latent topic models, such as Latent Dirichlet Allocation (LDA), which have been successfully used to discover individual activities, do not discover group activities. In this paper we propose a method to discover group activities along with individual activities. We use a two layer latent structure where a latent variable is used to discover correlation of individual activities as a group activity using multinomial distribution. Each individual activity is in turn represented as a distribution over local visual features. We use a Gibbs sampling-based algorithm to jointly infer the individual and group activities. Our method can summarize not only the individual activities but also the common group activities in a video. We demonstrate the strength of our method by discovering activities and the salient correlation amongst them in real life videos of crowded public places.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Blei, D.M., Ng, A.Y., Jordan, M.I., Lafferty, J.: Latent Dirichlet allocation. J. Mach. Learn. Res. 3, 993–1022 (2003)
Brand, M., Oliver, N., Pentland, A.: Coupled hidden Markov models for complex action recognition. In: CVPR (1997)
Choudhary, A., Pal, M., Banerjee, S., Chaudhury, S.: Unusual activity analysis using video epitomes and PLSA. In: ICVGIP, pp. 390–397 (2008)
Dollár, P., Rabaud, V., Cottrell, G., Belongie, S.: Behavior recognition via sparse spatio-temporal features. In: VS-PETS, pp. 65–72 (2005)
Faruquie, T.A., Kalra, P.K., Banerjee, S.: Time based activity inference using latent Dirichlet allocation. In: BMVC (2009)
Faruquie, T.A., Banerjee, S., Kalra, P.K.: Unsupervised discovery of activity correlations using latent topic models. In: Proceedings of the Seventh Indian Conference on Computer Vision, Graphics and Image Processing, pp. 25–32 (2010)
Hoffmann, T.: Probabilistic latent semantic analysis. In: SIGIR, pp. 50–57 (1999)
Hongeng, S., Nevatia, R.: Multi-agent event recognition. In: ICCV, pp. 84–93 (2001)
Hospedales, T., Gong, S., Xiang, T.: A Markov clustering topic model for mining behaviour in video. In: ICCV, pp. 1165–1172 (2009)
Hu, W., Xiao, X., Fu, Z., Xie, D., Tan, T., Maybank, S.: A system for learning statistical motion patterns. IEEE Trans. Pattern Anal. Mach. Intell. 28, 1450–1464 (2006)
Kuettel, D., Breitenstein, M.D., Gool, L.V., Ferrari, V.: What’s going on? discovering spatio-temporal dependencies in dynamic scenes. In: IEEE Conference on Computer Vision and Pattern Recognition, June (2010)
Li, J., Gong, S., Xiang, T.: Global behaviour inference using probabilistic latent semantic analysis. In: BMVC (2008)
Neal, R.: Markov chain sampling methods for Dirichlet process mixture models. J. Comput. Graph. Stat. 249–265 (2000)
Niebles, J., Wang, H., Li, F.: Unsupervised learning of human action categories using spatial-temporal words. Int. J. Comput. Vis. 79(3) (2008)
Ronning, G.: Maximum likelihood estimation of Dirichlet distributions. J. Stat. Comput. Simul. 32(4), 215–221 (1989)
Savarese, S., Pozo, A.D., Niebles, J.C., Li, F.F.: Spatial temporal correlations for unsupervised action classification. In: IEEE Workshop on Motion Video Compute, pp. 1–8 (2008)
Teh, Y., Jordon, M., Beal, M., Blei, D.: Hierarchical Dirichlet process. J. Am. Stat. Assoc. 1566–1581 (2006)
Vitaladevuni, S., Kellokumpu, V., Davis, L.: Action recognition using ballistic dynamics. In: CVPR (2008)
Wang, X., Tieu, K., Grimson, E.: Learning semantic scene models by trajectory analysis. In: ECCV, pp. 110–123 (2006)
Wang, X., Ma, X., Grimson, E.: Unsupervised activity perception by hierarchical Bayesian models. In: Proc. CVPR (2007)
Wang, X., Ma, X., Grimson, W.: Unsupervised activity perception in crowded and complicated scenes using hierarchical Bayesian models. IEEE Trans. Pattern Anal. Mach. Intell. 31(3), 539–555 (2009)
Wang, X., Tieu, K., Grimson, W.E.L.: Correspondence-free activity analysis and scene modeling in multiple camera views. IEEE Trans. Pattern Anal. Mach. Intell. 32, 56–71 (2010)
Wang, Y., Mori, G.: Human action recognition by semilatent topic models. IEEE Trans. Pattern Anal. Mach. Intell. 1762–1774 (2009)
Zhang, J., Gong, S.: Action categorization by structural probabilistic latent semantic analysis. Comput. Vis. Image Understan. (2010)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Faruquie, T.A., Banerjee, S. & Kalra, P. Latent topic model-based group activity discovery. Vis Comput 27, 1071–1082 (2011). https://doi.org/10.1007/s00371-011-0652-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00371-011-0652-1