Variational Inference for Background Subtraction in Infrared Imagery

TL;DR

This paper introduces a Bayesian Gaussian mixture model for infrared background subtraction that automatically estimates model parameters, adapts to changing conditions, and achieves high accuracy with real-time computational efficiency.

Contribution

It presents an analytically derived, sampling-free Bayesian approach for background modeling in infrared imagery, enabling automatic component estimation and dynamic adaptation.

Findings

Outperforms existing methods in precision and recall

Maintains real-time computational efficiency

Effectively adapts to changing environmental conditions

Abstract

We propose a Gaussian mixture model for background subtraction in infrared imagery. Following a Bayesian approach, our method automatically estimates the number of Gaussian components as well as their parameters, while simultaneously it avoids over/under fitting. The equations for estimating model parameters are analytically derived and thus our method does not require any sampling algorithm that is computationally and memory inefficient. The pixel density estimate is followed by an efficient and highly accurate updating mechanism, which permits our system to be automatically adapted to dynamically changing operation conditions. Experimental results and comparisons with other methods show that our method outperforms, in terms of precision and recall, while at the same time it keeps computational cost suitable for real-time applications.