Articulated Video Object Extraction by the Combination of Spatial and Temporal Segmentation

J. Gao and N. Thakoor (USA)


Optical field, motion estimation, motion segmentation, camera motion, displaced frame difference


Generally moving object extraction schemes rely on ei ther optical flow or frame difference. Though optical flow based methods can deal with moving camera situation with ease, the inconsistency at object boundaries causes extrac tion with inaccurate object boundaries. While frame differ ence approach, such as three frame difference, yields ac curate object boundaries, it usually cannot deal with mov ing camera. In this paper we present a novel technique for automatic object extraction captured by a moving camera. Our approach blends motion information from both opti cal flow and frame difference, and combines spatial region segementation for smooth object boundary detection. As a reprentative example of articulated object, experimental results of human extraction are presented.

Important Links:

Go Back