Waterpixels: Superpixels based on the watershed transformation

Abstract : Many sophisticated segmentation algorithms rely on a first low-level segmentation step where an image is partitioned into homogeneous regions with enforced compactness and adherence to object boundaries. These regions are called " su-perpixels ". While the marker controlled watershed transformation should in principle be well suited for this type of application , it has never been seriously tested in this setup, and comparisons to other methods were not made with the best possible settings. Here, we provide a scheme for applying the watershed transform for superpixel generation, where we use a spatially regularized gradient to achieve a tunable trade-off between superpixel regularity and adherence to object boundaries. We quantitatively evaluate our method on the Berkeley segmenta-tion database and show that we achieve comparable results to a previously published state-of-the art algorithm, while avoiding some of the arbitrary postprocessing steps the latter requires .
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [10 references]  Display  Hide  Download

https://hal-mines-paristech.archives-ouvertes.fr/hal-01139797
Contributor : Etienne Decencière <>
Submitted on : Tuesday, April 7, 2015 - 9:52:25 AM
Last modification on : Wednesday, February 13, 2019 - 2:24:02 PM
Document(s) archivé(s) le : Tuesday, April 18, 2017 - 11:56:17 AM

File

2014.icip.machairas.al.waterpi...
Files produced by the author(s)

Identifiers

Citation

Vaïa Machairas, Etienne Decencière, Thomas Walter. Waterpixels: Superpixels based on the watershed transformation. IEEE International Conference On Image Processing, Oct 2014, Paris, France. ⟨10.1109/ICIP.2014.7025882⟩. ⟨hal-01139797⟩

Share

Metrics

Record views

493

Files downloads

1406