8-2 映像の被写体領域抽出を目的とした時空間領域分割手法の検討(第8部門コンピュータビジョン,ヒューマンインタフェース,その他)

抄録

We propose a new video segmentation method for the sake of object extraction in video compositing. Our method is based on mean shift segmentation, and has improvements to meet two requirements of segmentation specific to make object extraction effective. We demonstrate the effectiveness of our method by extraction example.