Background Modeling Method Based on 3D Shape Reconstruction Technology

Abstract

In this research, we present a novel dynamic background modeling method based on reconstructed 3D shapes, which can solve background modeling problems of multi-camera in real-time. While 3D shape reconstruction is a popular technology widely used for detecting, tracking or identifying various objects, little effort has been made in applying this useful method to background subtraction. In this work we propose an approach to using 3D shape reconstruction technology to develop a novel decision making mechanism for background image updating. This 3D shape reconstruction based background subtraction method is adaptive to changes in illumination, capable of handling sudden illumination changes as well as complex dynamic scenes efficiently. DOI: http://dx.doi.org/10.11591/telkomnika.v11i4.2387