Segmentation of Moving Object In Video Using Background Registration and GMEC

Emerging multimedia applications demand content-based video processing. Content based video retrieval or concept detection systems require video to be segmented in to objects. A large number of video object segmentation algorithms have been developed such as semiautomatic and automatic. Semiautomatic methods requires the human intervention and also not suitable for some applications. Many applications require automatic segmentation but still there is lot of scope for the improvement. The proposed work aims at contributing to identify the lacunas that are present in the existing segmentation system and also to give the possible solutions to overcome those lacunas so that the precise and effective video segmentation system can be developed. The proposed system deals with the issue of automatic video segmentation system such as uncovered background, Temporary poses and Global motion of background.