What to Show? - Automatic Stream Selection among Multiple Sensors

Rémi Emonet, E. Oberzaucher, J.-M. Odobez

2014

Abstract

The installation of surveillance networks has been growing exponentially in the last decade. In practice, videos from large surveillance networks are almost never watched, and it is frequent to see surveillance video wall monitors showing empty scenes. There is thus a need to design methods to continuously select streams to be shown to human operators. This paper addresses this issue and make three main contributions: it introduces and investigates, for the first time in the literature, the live stream selection task; based on the theory of social attention, it formalizes a way of obtaining some ground truth for the task and hence a way of evaluating stream selection algorithms; and finally, it proposes a two-step approach to solve this task and compares different approaches for interestingness rating using our framework. Experiments conducted on 9 cameras from a metro station and 5 hours of data randomly selected over one week show that, while complex unsupervised activity modeling algorithms achieve good performance, simpler approaches based on amount of motion perform almost as well for this type of indoor setting.

References

  1. Atrey, P. K., El Saddik, A., and Kankanhalli, M. S. (2011). Effective multimedia surveillance using a humancentric approach. Multimedia Tools and Applications, 51(2):697-721.
  2. Boiman, O. and Irani, M. (2007). Detecting irregularities in images and in video. International Journal of Computer Vision, 74(1):17-31.
  3. Emonet, R., Varadarajan, J., and Odobez, J.-M. (2011). Multi-camera Open Space Human Activity Discovery for Anomaly Detection. In 8th IEEE Int. Conf. on Advanced Video and Signal-Based Surveillancei (AVSS), page 6.
  4. Hua, D., Zhang, X., Yin, J., Zheng, V. W., and Yang, Q. (2009). Abnormal activity recognition based on hdphmm models. In Int. Joint Conference in Artificial Intelligence.
  5. Varadarajan, J. and Odobez, J. (2009). Topic models for scene analysis and abnormality detection. In ICCV12th International Workshop on Visual Surveillance.
  6. Zelniker, E., Gong, S., and Xiang, T. (2008). Global abnormal behaviour detection using a network of cctv cameras. In IEEE International Workshop on Visual Surveillance.
  7. Zhong, H., Shi, J., and Visontai, M. (2004). Detecting unusual activity in video. In IEEE Conference on Computer Vision and Pattern Recognition, volume 2, pages 819-826, Washington, DC.
Download


Paper Citation


in Harvard Style

Emonet R., Oberzaucher E. and Odobez J. (2014). What to Show? - Automatic Stream Selection among Multiple Sensors . In Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014) ISBN 978-989-758-004-8, pages 433-440. DOI: 10.5220/0004688504330440


in Bibtex Style

@conference{visapp14,
author={Rémi Emonet and E. Oberzaucher and J.-M. Odobez},
title={What to Show? - Automatic Stream Selection among Multiple Sensors},
booktitle={Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)},
year={2014},
pages={433-440},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004688504330440},
isbn={978-989-758-004-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)
TI - What to Show? - Automatic Stream Selection among Multiple Sensors
SN - 978-989-758-004-8
AU - Emonet R.
AU - Oberzaucher E.
AU - Odobez J.
PY - 2014
SP - 433
EP - 440
DO - 10.5220/0004688504330440