Abstract：This paper prroposes a novel semiautomatic method for semantic video object extraction. The proposed method aims at improving the semiautomatic segmentation by integrating various semantic properties of the video. Object region in the first frame is extraded by a user input polygon. Then, a tracking scheme based on backward block-marching is used to automatically extract object region in the rest of frames for general objects. Translational rigid objects and slowly changing objects are specially considered in the tracking. Finally, a mask refinement algorithm based on peer group is applied to extract accurate object boundary and insure reliaible tracking. Accurate and consistent results are obtained in the experiments by the proposed method.
[1] Ju Guo,Jongwon Kim,Kuo C C J.An interactive object segmentation system for MPEG video[A].Proceedings of ICIP 99[C].ICIP,1999.[2] Dong Kwon Park,Ho Seek Yoon,Chee Sun Won.Fast object tracking in digital video[J].IEEE Trans.on Consumer Electronics,2000,46(3):785-790.[3] Grinias I,Tziritas G.A semi-automatic seeded region growing algorithm for video object localization and tracking[J].Signal Processing:Image Communication,2001.[4] Chuang Gu,Ming-Chieh Lee.Semiautomatic segmentation and tracking of semantic video objects[J].IEEE Trans.on Circuits and Systems for Video Technology,1998,8(5):572-584.[5] Comaniciu D,Ramesh V,Meer P.The variable bandwidth mean shift and data-driven scale selection[A].Proceedings of ICCV01[C].ICCV,2001.[6] Yining Deng,Manjunath B S.Unsupervised segmentation of color-texture regions in images and video[J].IEEE Trans.on Pattern Analysis and Machine Intelligence,2001,23(8):800-810.[7] Yining Deng,Kenney C,Moore M S,Manjunath B S.Peer group filtering and perceptual color image quantization[A].Proceedings of ISCAS99[C].ISCAS,1999.4:21-24.
2002, Vol. 30 
