Dynamic Hierarchical Dirichlet Process for Abnormal Behaviour Detection in Video

TL;DR

This paper introduces a dynamic Hierarchical Dirichlet Process model with online inference algorithms for real-time abnormal behaviour detection in videos, improving classification accuracy over static models.

Contribution

It develops a novel dynamic HDP model with incremental Gibbs sampling for sequential data, specifically applied to video abnormality detection.

Findings

Dynamic HDP outperforms static HDP in abnormal behaviour detection

Online Gibbs sampling enables real-time processing of video data

The proposed method improves classification accuracy on synthetic and real datasets

Abstract

This paper proposes a novel dynamic Hierarchical Dirichlet Process topic model that considers the dependence between successive observations. Conventional posterior inference algorithms for this kind of models require processing of the whole data through several passes. It is computationally intractable for massive or sequential data. We design the batch and online inference algorithms, based on the Gibbs sampling, for the proposed model. It allows to process sequential data, incrementally updating the model by a new observation. The model is applied to abnormal behaviour detection in video sequences. A new abnormality measure is proposed for decision making. The proposed method is compared with the method based on the non- dynamic Hierarchical Dirichlet Process, for which we also derive the online Gibbs sampler and the abnormality measure. The results with synthetic and real data show that the consideration of the dynamics in a topic model improves the classification performance for abnormal behaviour detection.