Efficiently Segmenting Images with Dominant Sets
Dominant sets are a new graph-theoretic concept that has proven to be relevant in clustering as well as image segmentation problems. However, due to the computational loads of this approach, applications to large problems such as high resolution imagery have been unfeasible. In this paper we provide a method that substantially reduces the computational burden of the dominant set framework, making it possible to apply it to very large grouping problems. Our approach is based on a heuristic technique that allows one to obtain the complete grouping solution using only a small number of samples.
KeywordsImage Segmentation Similarity Graph Evolutionary Game Theory Heuristic Technique Replicator Equation
Unable to display preview. Download preview PDF.
- 1.Forsyth, D., Ponce, J.: Computer Vision: A Modern Approach. Prentice-Hall, Englewood Cliffs (2002)Google Scholar
- 3.Martin, D., Fowlkes, C., Tal, D., Malik, J.: A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In: Proc. of the IEEE Int. Conf. on Computer Vision, vol. 2, pp. 416–424 (2001)Google Scholar
- 5.Pavan, M., Pelillo, M.: A new graph-theoretic approach to clustering and segmentation. In: Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol. 1, pp. 145–152 (2003)Google Scholar
- 6.Pavan, M., Pelillo, M.: Unsupervised texture segmentation by dominant sets and game dynamics. In: Proc. IEEE Int. Conf. on Image Analysis and Processing, pp. 302–307 (2003)Google Scholar