New Trends In Computer Networks
Spanning a wide spectrum of topics in computer networks, this book presents a selective collection of papers from the 20th International Symposium on Computer and Information Sciences, held in Istanbul, Turkey.
TABLE OF CONTENTS New Trends In Computer Networks
Video Segmentation with Block Based Change Detection Using Numerous Preceding Image Frames
O. URHAN, S. ERT RK
Electronics and Telecommunications Engineering Department, Veziroglu Campus, University of Kocaeli, 41040, Kocaeli, TR. & LATARUB Research Centre, University of Kocaeli, Kocaeli, 41380, TR.
This paper presents a robust video segmentation method particularly aimed at videophone and videoconference type video sequences containing slow moving objects. Classical change detection approaches based only on the frame difference of two successive image frames usually result in incorrect segmentations for video sequences containing slow motions or temporary poses. The video segmentation approach presented in this paper carries out block based change detection using numerous preceding image frames to accomplish successful segmentation in the presence of slow motions and temporary poses.
1. Introduction
Segmentation of digital video plays an important role for content based multimedia applications. The new video coding standard MPEG-4 for instance relies on the decomposition of image sequence frames into semantically meaningful video objects (VO) to provide content based functionalities. These content based functionalities allow separate coding and encoding of objects and permit the manipulation of the original scene by simple operations on the bit stream [1]. In this case, each object in an image frame is represented as a video object plane (VOP) with contour, motion and texture information being coded in a separate VOP-layer in the MPEG4 visual coding standard.
Moving object segmentation techniques can be separated into motion-based methods and spatio-temporal methods [2]. Motion based methods can be classified as 2D [3-7] and 3D techniques. 2D motion based methods have usually a relatively low computational complexity...
Copyright Imperial College Press 2005 under license agreement with Books24x7
