Dilip Mathew Thomas

This project reviewed different techniques for detecting anomalies in videos using unsupervised machine learning techniques. We explored the use of reconstruction, predictive, and deep learning-based generative models for this project.

These reconstruction-based models build representations that minimize the reconstruction error of training samples from the normal distribution. Spatio-temporal predictive models consider the correlation by viewing videos as a spatiotemporal time series and learning representations that minimize the prediction error on spatiotemporal sequences. The generative models learn to generate samples from the training distribution while minimizing the reconstruction error and the distance between generated and trained distribution. Each of these methods focuses on learning prior information useful for constructing the representation for the video anomaly detection task.