Motion objects segmentation based on structural similarity background modelling

What is it about?

It is important to efficiently segment motion objects from video in computer vision applications. A novel foreground segmentation approach has been developed based on structural similarity background modelling, which responds quickly to sudden illumination changes and dynamic background.

Featured Image

Why is it important?

The contributions of this paper are as follows. Firstly, structural similarity based background modeling has been proposed to fast respond to sudden illumination changes and dynamic background. Secondly, an adaptive updating background strategy has been developed in which both the structural similarity map and the environmental variation parameters are taken as the dynamic feedback controllers. Finally, a multi-modal features fusion strategy has been proposed to segment foregrounds in a dynamic cluttered scene without any hypothesis for the scenario contents in advance.