Change Detection for Object Segmentation
Object segmentation based on motion or change detection is the first step in many smart camera applications, such as video surveillance, human computer interaction, and immersive gaming. Efficient and accurate techniques are required to detect and label the changes in the observed scene. Different change detection techniques can be employed for moving camera and static camera conditions. If the camera moves, change detection aims to recognize coherent and non-coherent moving areas. The former correspond to background areas, the latter to moving objects. If the camera is static, the goal of change detection is to recognize moving objects (foreground) and the static background. In this chapter, we focus on change detection applications when the background is static. The methods discussed here are applicable in the case of a moving smart camera after global motion compensation. Change detection algorithms proposed in the literature are discussed based on four building blocks of a proposed general model.
KeywordsReference Frame Change Detection Previous Frame Object Segmentation Binary Mask
Unable to display preview. Download preview PDF.