Figure-Ground Segmentation using Tabu Search
|
Abstract:
Many computer vision problems, such as figureground segmentation, simultaneous fitting of curves, selection of an optimal set of geometric primitives, can be formulated naturally as discrete optimization problems. The statement of these problems is relatively easy, but to find techniques that efficiently solve them constitutes a major challenge. In this paper we focus on figure-ground segmentation. We present a Tabu search strategy which is able to solve the discrete optimization problem associated with figure-ground segmentation in a very efficient way. The resulting deterministic algorithm outperforms the currently fastest known algorithm to solve this problem (mean field annealing) by two orders of magnitude in speed and in addition it consistently finds better optima. 1
Citations
| 425 | Neural computation of decisions in optimization problems – Hopfield, Tank - 1985 |
| 375 | Tabu Search – Glover - 1989 |
| 75 | A common framework for image segmentation – Geiger, Yuille - 1991 |
| 57 | Segmentation of range images as the search for geometric parametric models – Leonardis, Gupta, et al. - 1995 |
| 40 | Figure-ground discrimination: A combinatorial optimization approach – Herault, Horaud - 1993 |
| 28 | Part Segmentation for Object Recognition – Pentland - 1989 |
| 10 | Stochastic relaxation, Gibbs distributions and the Baysian restoration of images – Geman, Geman - 1984 |
| 7 | Simultaneous fitting of several planes to point sets using neural networks – Kamgar-Parsi, Kamgar-Parsi, et al. - 1990 |
| 2 | Tabu search: an efficient solution to discrete optimization problems – Stricker, Leonardis - 1995 |
