Video Object Segmentation of Still Background
Yun Liu · Journal of Qingdao University of Science and Technology · 2004
In this paper, according to the characteristics of the foreground and the background, we propose a novel video object segmentation algorithm, which algorithm is a Bayes decision-making theory with least risk based on still background. Experiments on the platform of TMS320C6711 IDK(Imaging Developer's Kit) of Texas Instruments implies that this algorithm can segment video object and meet needs of applications of MPEG-4.