Superpixels Optimized by Color and Shape

  • Vitaliy KurlinEmail author
  • Donald Harvey
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10746)


Image over-segmentation is formalized as the approximation problem when a large image is segmented into a small number of connected superpixels with best fitting colors. The approximation quality is measured by the energy whose main term is the sum of squared color deviations over all pixels and a regularizer encourages round shapes. The first novelty is the coarse initialization of a non-uniform superpixel mesh based on selecting most persistent edge segments. The second novelty is the scale-invariant regularizer based on the isoperimetric quotient. The third novelty is the improved coarse-to-fine optimization where local moves are organized according to their energy improvements. The algorithm beats the state-of-the-art on the objective reconstruction error and performs similarly to other superpixels on the benchmarks of BSD500.


Superpixel Segmentation Approximation Boundary Recall Reconstruction error Energy minimization Coarse-to-fine optimization 


  1. 1.
    Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P., Süsstrunk, S.: SLIC superpixels compared to the state-of-the-art. Trans. PAMI 34, 2274–2282 (2012)CrossRefGoogle Scholar
  2. 2.
    Van de Bergh, M., Boix, X., Roig, G., Van Gool, L.: SEEDS: superpixels extracted via energy-driven sampling. Int. J. Comput. Vis. 111, 298–314 (2015)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Yao, J., Boben, M., Fidler, S., Urtasun, R.: Real-time coarse-to-fine topologically preserving segmentation. In: Proceedings of CVPR, pp. 216–225 (2015)Google Scholar
  4. 4.
    Arbelaez, P., Maire, M., Fowlkes, C., Malik, J.: Contour detection and hierarchical image segmenetaton. Trans. PAMI 33, 898–916 (2011)CrossRefGoogle Scholar
  5. 5.
    Stutz, D., Hermans, A., Leibe, B.: Superpixels: an evaluation of the state-of-the-art. Comput. Vis. Image Underst. 166, 1–27 (2017)CrossRefGoogle Scholar
  6. 6.
    Shi, J., Malik, J.: Normalized cuts and image segmentation. Trans. PAMI 22, 888–905 (2000)CrossRefGoogle Scholar
  7. 7.
    Liu, M.Y., Tuzel, O., Ramalingam, S., Chellappa, R.: Entropy rate superpixel segmentation. In: Proceedings of CVPR, pp. 2097–2104 (2011)Google Scholar
  8. 8.
    Conrad, C., Mertz, M., Mester, R.: Contour-relaxed superpixels. In: Heyden, A., Kahl, F., Olsson, C., Oskarsson, M., Tai, X.C. (eds.) Energy Minimization Methods in Computer Vision and Pattern Recognition. LNCS, vol. 8081, pp. 280–293. Springer, Heidelberg (2013). CrossRefGoogle Scholar
  9. 9.
    Duan, L., Lafarge, F.: Image partitioning into convex polygons. In: Proceedings of CVPR, pp. 3119–3127 (2015)Google Scholar
  10. 10.
    Forsythe, J., Kurlin, V., Fitzgibbon, A.: Resolution-independent superpixels based on convex constrained meshes without small angles. In: Bebis, G., et al. (eds.) ISVC 2016. LNCS, vol. 10072, pp. 223–233. Springer, Cham (2016). Google Scholar
  11. 11.
    Chernov, A., Kurlin, V.: Reconstructing persistent graph structures from noisy images. Image-A 3, 19–22 (2013)Google Scholar
  12. 12.
    Kurlin, V.: Auto-completion of contours in sketches, maps and sparse 2D images based on topological persistence. In: Proceedings of CTIC, pp. 594–601 (2014)Google Scholar
  13. 13.
    Kurlin, V.: A fast persistence-based segmentation of noisy 2D clouds with provable guarantees. Pattern Recogn. Lett. 83, 3–12 (2016)CrossRefGoogle Scholar
  14. 14.
    Kurlin, V.: A one-dimensional homologically persistent skeleton of a point cloud in any metric space. Comput. Graph. Forum 34, 253–262 (2015)CrossRefGoogle Scholar
  15. 15.
    Kurlin, V.: A homologically persistent skeleton is a fast and robust descriptor of interest points in 2D images. In: Azzopardi, G., Petkov, N. (eds.) CAIP 2015. LNCS, vol. 9256, pp. 606–617. Springer, Cham (2015). CrossRefGoogle Scholar
  16. 16.
    Kurlin, V.: A fast and robust algorithm to count topologically persistent holes in noisy clouds. In: Proceedings of CVPR, pp. 1458–1463 (2014)Google Scholar
  17. 17.
    Schick, A., Fischer, M., Stifelhagen, R.: Measuring and evaluating the compactness of superpixels. In: Proceedings of ICPR, pp. 930–934 (2012)Google Scholar
  18. 18.
    Neubert, P., Protzel, P.: Compact watershed and preemptive SLIC: on improving trade-offs of superpixel segmentation algorithms. In: proceedings of the ICPR, pp. 996–1001 (2014)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversity of LiverpoolLiverpoolUK
  2. 2.Main Yard StudiosLondonUK

Personalised recommendations