Can DMD obtain a Scene Background in Color?

TL;DR

This paper extends the Dynamic Mode Decomposition (DMD) method to extract scene backgrounds in color, demonstrating its effectiveness on a standard dataset for applications like surveillance and photography.

Contribution

The study proposes a novel technique to apply DMD directly in the color domain for background modeling, improving upon previous grayscale-based methods.

Findings

DMD successfully extracts colored backgrounds in videos.

The method performs well on the SBI dataset.

Qualitative and quantitative results confirm effectiveness.

Abstract

A background model describes a scene without any foreground objects and has a number of applications, ranging from video surveillance to computational photography. Recent studies have introduced the method of Dynamic Mode Decomposition (DMD) for robustly separating video frames into a background model and foreground components. While the method introduced operates by converting color images to grayscale, we in this study propose a technique to obtain the background model in the color domain. The effectiveness of our technique is demonstrated using a publicly available Scene Background Initialisation (SBI) dataset. Our results both qualitatively and quantitatively show that DMD can successfully obtain a colored background model.