Darren Butler, Sridha Sridharan, V. Michael Bove, Jr.

Abstract

Automatic analysis of digital video scenes often requires the segmentation of moving objects from the background. Historically, algorithms developed for this purpose have been restricted to small frame sizes, low frame rates or offline processing. The simplest approach involves subtracting the current frame from the known background. However, as the background is unknown, the key is how to learn and model it. This paper proposes a new algorithm that represents each pixel in the frame by a group of clusters. The clusters are ordered according the likelihood that they model the background and are adapted to deal with background and lighting variations. Incoming pixels are matched against the corresponding cluster group and are classified according to whether the matching cluster is considered part of the background. The algorithm has been subjectively evaluated against three other techniques. It demonstrated equal or better segmentation than the other techniques and proved capable of processing 320 × 240 video at 28 fps, excluding post-processing.