Authors:
Henrique M. Gonçalves
1
;
Gustavo J. Q. de Vasconcelos
1
;
Paola R. R. Rangel
2
;
Murilo Carvalho
3
;
Nathaly L. Archilha
4
and
Thiago V. Spina
4
Affiliations:
1
Brazilian Synchrotron Light Laboratory, Brazilian Center for Research in Energy and Materials, Campinas, SP, Brazil, Institute of Computing, University of Campinas, Campinas, SP and Brazil
;
2
Brazilian Synchrotron Light Laboratory, Brazilian Center for Research in Energy and Materials, Campinas, SP, Brazil, Institute of Geosciences, University of Campinas, Campinas, SP and Brazil
;
3
Brazilian Bioscences National Laboratory, Brazilian Center for Research in Energy and Materials, Campinas, SP and Brazil
;
4
Brazilian Synchrotron Light Laboratory, Brazilian Center for Research in Energy and Materials, Campinas, SP and Brazil
Keyword(s):
Image Foresting Transform, GPU, Watershed, Image Segmentation, Superpixels.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Image and Video Analysis
;
Segmentation and Grouping
Abstract:
We propose a GPU-based version of the Image Foresting Transform by Seed Competition (IFT-SC) operator and instantiate it to produce compact watershed-based superpixels (Waterpixels). Superpixels are usually applied as a pre-processing step to reduce the amount of processed data to perform object segmentation. However, recent advances in image acquisition techniques can easily produce 3D images with billions of voxels in roughly 1 second, making the time necessary to compute Waterpixels using the CPU-version of the IFT-SC quickly escalate. We aim to address this fundamental issue, since the efficiency of the entire object segmentation methodology may be hindered by the initial process of estimating superpixels. We demonstrate that our CUDA-based version of the sequential IFT-SC operator can speed up computation by a factor of up to 180x for 2D images, with consistent optimum-path forests without requiring additional CPU post-processing.