Automatic Segmentation of Video Object under Static Background

Segmentation of semantic video object planes (VOPpsilas) from video sequence is a key to the standard MPEG-4 with content-based video coding. In our paper, we propose an automatic segmentation algorithm under static background. The algorithm can extract accurate video objects from slow moving video sequences. The initial two coarse masks are first obtained based on frame difference and motion detection, then the two coarse masks are further processed by a lot of post-processing algorithms. Finally, the final mask guides the extraction. The final mask is got through logic ldquoANDrdquo of the two corrected masks. The proceeding results demonstrate the good performance of the algorithm.