Unsupervised Synthesis of Anomalies in Video: Transforming the Normal

Abhishek Joshi
Vinay P. Namboodiri
Indian Institute of Technology Kanpur




Abstract

Abnormal activity recognition requires detection of occurrence of anomalous events that suffer from a severe imbalance in data. In a video, normal is used to describe activities that conform to usual events while the irregular events which do not conform to the normal are referred to as abnormal. It is far more common to observe normal data than to obtain abnormal data in visual surveillance. In this paper, we propose an approach where we can obtain abnormal data by transforming normal data. This is a challenging task that is solved through a multi-stage pipeline approach. We utilize a number of techniques from unsupervised segmentation in order to synthesize new samples of data that are transformed from an existing set of normal examples. An incrementally trained Bayesian convolutional neural network (CNN) is used to carefully select the set of abnormal samples that can be added. Finally through this synthesis approach we obtain a comparable set of abnormal samples that can be used for training the CNN for the classification of normal vs abnormal samples. We show that this method generalizes to multiple settings by evaluating it on two real world datasets and achieves improved performance over other probabilistic techniques that have been used in the past for this task.



Demo Video


[Watch Demo]


Synthesized Samples





[ Paper ] [ arXiv ]

A. Joshi, V. P. Namboodiri
Unsupervised Synthesis of Anomalies in Video: Transforming the Normal


[Bibtex]



Acknowledgements

We acknowledge the travel support from Microsoft Research India.
We thank the members of DelTA Lab for helpful discussions.



Citation

If you find this useful for your research, please use the following.

@InProceedings{abhishekjoshi_2019_IJCNN,
   title={Unsupervised Synthesis of Anomalies in Video: Transforming the Normal},
   author={Joshi, Abhishek and Namboodiri, Vinay P.},
   booktitle={International Joint Conference on Neural Networks (IJCNN)},
   year={2019}
}