Spatio-temporal Segmentation Using Dominant Sets
Pairwise data clustering techniques are gaining increasing popularity over traditional, feature-based central grouping techniques. These approaches have proven very powerful when applied to image-segmentation problems. However, they are computationally too demanding to be applied to video-segmentaton problems due to their scaling behavior with the quantity of data. On a dataset containing N examples, the number of potential comparisons scales with O(N 2), thereby rendering the approaches unfeasible for problems involving very large data sets. It is therefore of primary importance to develop strategies to reduce the number of comparisons required by subsampling the data and extending the grouping to out-of-sample points after the clustering process has taken place. In this paper we present an approach to out-of-sample clustering based on the dominant set framework  and apply it to video segmentation. The method is compared against two recent pairwise clustering algorithms which provide out-of-sample extensions: the Nyström method , and the minimal-shift embedding approach . Our results show that our approach performs comparably against the competition in terms of quality of the segmentation, being, however, much faster.
Unable to display preview. Download preview PDF.
- 1.Bengio, Y., Paiement, J.-F., Vincent, P., Delalleau, O., Le Roux, N., Ouimet, M.: Out-of-sample extensions for LLE, isomap, MDS, eigenmaps, and spectral clustering. In: Advances in Neural Information Processing Systems, vol. 16. MIT Press, Cambridge (2004)Google Scholar
- 4.Hamid, R., et al.: Detection and explanation of anomalous activities: representing activities as bags of event n-grams. In: CVPR 2005 (2005) (to appear)Google Scholar
- 9.Ng, A.Y., Jordan, M.I., Weiss, Y.: On spectral clustering: Analysis and an algorithm. In: Advances in Neural Information Processing Systems, vol. 14, pp. 849–856. MIT Press, Cambridge (2002)Google Scholar
- 10.Pavan, M., Pelillo, M.: A new graph-theoretic approach to clustering and segmentation. In: Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 145–152 (2003)Google Scholar
- 11.Pavan, M., Pelillo, M.: Unsupervised texture segmentation by dominant sets and game dynamics. In: Proc. 12th Int. Conf. on Image Analysis and Processing, pp. 302–307 (2003)Google Scholar
- 12.Pavan, M., Pelillo, M.: Efficient out-of-sample extension of dominant-set clusters. In: Advances in Neural Information Processing Systems, vol. 17, pp. 1057–1064. MIT Press, Cambridge (2005)Google Scholar
- 13.Perona, P., Freeman, W.: A factorization approach to grouping. In: Burkhardt, H.-J., Neumann, B. (eds.) ECCV 1998. LNCS, vol. 1406, pp. 655–670. Springer, Heidelberg (1998)Google Scholar
- 18.Weiss, Y.: Segmentation using eigenvectors: A unifying view. In: Proc. 7th Int. Conf. on Computer Vision, pp. 975–982 (1999)Google Scholar