Learning Methods for Dynamic Topic Modeling in Automated Behaviour Analysis

TL;DR

This paper introduces new learning algorithms for dynamic topic modeling in video activity analysis, enabling autonomous behavior detection and anomaly localization in large video datasets.

Contribution

It presents two novel algorithms based on expectation maximisation and variational Bayes, with theoretical derivations and empirical comparisons to existing methods.

Findings

The new algorithms outperform Gibbs sampling in accuracy and efficiency.

The framework effectively localizes anomalies in video data.

Applicable to transportation, security, and surveillance domains.

Abstract

Semi-supervised and unsupervised systems provide operators with invaluable support and can tremendously reduce the operators load. In the light of the necessity to process large volumes of video data and provide autonomous decisions, this work proposes new learning algorithms for activity analysis in video. The activities and behaviours are described by a dynamic topic model. Two novel learning algorithms based on the expectation maximisation approach and variational Bayes inference are proposed. Theoretical derivations of the posterior of model parameters are given. The designed learning algorithms are compared with the Gibbs sampling inference scheme introduced earlier in the literature. A detailed comparison of the learning algorithms is presented on real video data. We also propose an anomaly localisation procedure, elegantly embedded in the topic modeling framework. The proposed framework can be applied to a number of areas, including transportation systems, security and surveillance.