FFT snake: a robust and efficient method for the segmentation of arbitrarily shaped objects in image sequences

A robust and efficient algorithm for segmenting arbitrarily shaped objects in images, which is called FFT snake, is proposed in this paper. A low-pass filter with the fast Fourier transform (FFT) of the curve as theoretic internal force is first introduced to smooth the contours. In real algorithm, it is composed of the curves trimming and crossing chains cutting. At last the contours are evolved in the direction of normal vectors of the curve to match the feature-map. The algorithm is then applied to the rapid video feedback on the motion for the real-time diving training. The results are highly encouraging to capture the contours of arbitrarily shaped objects for real-time tracking systems. We believe that FFT snake has wide uses in video compression, multimedia applications, and so on.

[1]  H. Saji,et al.  Deformable templates for tracking the facial components , 1997, Proceedings 6th IEEE International Workshop on Robot and Human Communication. RO-MAN'97 SENDAI.

[2]  Laurent D. Cohen,et al.  On active contour models and balloons , 1991, CVGIP Image Underst..

[3]  Jerry L. Prince,et al.  Snakes, shapes, and gradient vector flow , 1998, IEEE Trans. Image Process..

[4]  Terry E. Weymouth,et al.  Using Dynamic Programming For Minimizing The Energy Of Active Contours In The Presence Of Hard Constraints , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[5]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[6]  S. Osher,et al.  Level set methods: an overview and some recent results , 2001 .

[7]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[8]  Guillermo Sapiro,et al.  Geodesic Active Contours , 1995, International Journal of Computer Vision.