Abstract
Video data captured from a public space is typically characterised by highly imbalanced behaviour class distribution. Most of the captured data examples reflect normal behaviours. Unusual behaviours, either because of being rare or abnormal, only constituent a small portion in the observed data. Whilst an unsupervised learning based model can be constructed to detect unusual behaviours through a process of outlier detection, an outlier detection based model is fundamentally limited in a number of ways: (1) The model has difficulties in detecting subtle unusual behaviours. (2) The model does not exploit information from detected unusual behaviours. Image noise and errors in feature extraction could be mistaken as genuine unusual behaviours of interest, giving false alarms in detection. (3) A large amount of rare but benign behaviours give rise to false alarms in unusual behaviour detection. Human knowledge can be exploited in a different way to address this problem. Instead of giving supervision on the input to model learning by labelling the training data, human knowledge can be more effectively utilised by giving selective feedback to model output. In this chapter, we describe an active learning model for seeking human feedback on model selected queries. The query selection criteria are internal to the model rather than decided by the human observer. This is to ensure that not only the most relevant human feedback is sought, but also the model is not subject to human bias without data support.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In this book, we alternate the use of the terms ‘classification boundaries’ and ‘decision boundaries’.
- 2.
A symmetric Kullback–Leibler divergence \(\overline{\mathcal{K\!L}}( \varTheta~\|~\tilde{\varTheta})\) is required because in general, KL divergence is non-symmetric, where \(\mathcal{K\!L}( \varTheta~\|~\tilde{\varTheta}) \; \not{\!\!\equiv} \;\mathcal{K\!L}( \tilde{\varTheta}~\|~\varTheta)\). A non-symmetric KL divergence is not a true distance metric.
- 3.
References
Argamon-Engelson, S., Dagan, I.: Committee-based sample selection for probabilistic classifiers. J. Artif. Intell. Res. 11, 335–360 (1999)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, Berlin (2006)
Boiman, O., Irani, M.: Detecting irregularities in images and in video. Int. J. Comput. Vis. 74(1), 17–31 (2007)
Cover, T.M., Thomas, J.A.: Elements of Information Theory. Wiley, New York (2006)
Dagan, I., Engelson, S.: Committee-based sampling for training probabilistic classifiers. In: International Conference on Machine Learning, pp. 150–157 (1995)
Devroye, L.: Non-uniform Random Variate Generation. Springer, Berlin (1986)
Hand, D.J., Yu, K.: Idiot’s Bayes: Not so stupid after all? Int. Stat. Rev. 69(3), 385–398 (2001)
Ho, S., Wechsler, H.: Query by transduction. IEEE Trans. Pattern Anal. Mach. Intell. 30(9), 1557–1571 (2008)
Hospedales, T., Gong, S., **ang, T.: Finding rare classes: adapting generative and discriminative models in active learning. In: Pacific-Asia Conference on Knowledge Discovery and Data Mining, Shenzhen, China, May 2011
Kapoor, A., Horvitz, E.: On discarding, caching, and recalling samples in active learning. In: Uncertainty in Artificial Intelligence, pp. 209–216 (2007)
Li, J., Hospedales, T., Gong, S., **ang, T.: Learning rare behaviours. In: Asian Conference on Computer Vision, Queenstown, New Zealand, November 2010
Loy, C.C., **ang, T., Gong, S.: Stream-based active unusual event detection. In: Asian Conference on Computer Vision, Queenstown, New Zealand, November 2010
Lucas, B.D., Kanade, T.: An iterative image registration technique with an application to stereo vision. In: DARPA Image Understanding Workshop, April 1981, pp. 121–130 (1981)
McCallum, A., Nigam, K.: Employing EM in pool-based active learning for text classification. In: International Conference on Machine Learning, pp. 350–358 (1998)
Pelleg, D., Moore, A.: Active learning for anomaly and rare-category detection. In: Advances in Neural Information Processing Systems, pp. 1073–1080 (2004)
Rish, I.: An empirical study of the naive Bayes classifier. In: IJCAI Workshop on Empirical Methods in Artificial Intelligence, pp. 41–46 (2001)
Roy, N., McCallum, A.: Toward optimal active learning through sampling estimation of error reduction. In: International Conference on Machine Learning, pp. 441–448 (2001)
Settles, B.: Active learning literature survey. Technical report, University of Wisconsin Madison (2010)
Settles, B., Craven, M., Ray, S.: Multiple instance active learning. In: Advances in Neural Information Processing Systems, pp. 1289–1296 (2007)
Seung, H.S., Opper, M., Sompolinsky, H.: Query by committee. In: Conference on Learning Theory, pp. 287–294 (1992)
Tong, S., Koller, D.: Active learning for parameter estimation in Bayesian networks. In: Advances in Neural Information Processing Systems, pp. 647–653 (2000)
**ang, T., Gong, S.: Video behaviour profiling for anomaly detection. IEEE Trans. Pattern Anal. Mach. Intell. 30(5), 893–908 (2008)
Zelnik-Manor, L., Perona, P.: Self-tuning spectral clustering. In: Advances in Neural Information Processing Systems, pp. 1601–1608 (2004)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2011 Springer-Verlag London Limited
About this chapter
Cite this chapter
Gong, S., **ang, T. (2011). Man in the Loop. In: Visual Analysis of Behaviour. Springer, London. https://doi.org/10.1007/978-0-85729-670-2_12
Download citation
DOI: https://doi.org/10.1007/978-0-85729-670-2_12
Publisher Name: Springer, London
Print ISBN: 978-0-85729-669-6
Online ISBN: 978-0-85729-670-2
eBook Packages: Computer ScienceComputer Science (R0)