cudaIFT: 180x Faster Image Foresting Transform for Waterpixel Estimation using CUDA

Henrique M. Gonçalves, Gustavo J. Q. de Vasconcelos, Paola R. R. Rangel, Murilo Carvalho, Nathaly L. Archilha, Thiago V. Spina

2019

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.

Paper Citation

in Harvard Style

Gonçalves H., Q. de Vasconcelos G., Rangel P., Carvalho M., Archilha N. and Spina T. (2019). cudaIFT: 180x Faster Image Foresting Transform for Waterpixel Estimation using CUDA. In Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP; ISBN 978-989-758-354-4, SciTePress, pages 395-404. DOI: 10.5220/0007402703950404

in Bibtex Style

@conference{visapp19,
author={Henrique M. Gonçalves and Gustavo J. Q. de Vasconcelos and Paola R. R. Rangel and Murilo Carvalho and Nathaly L. Archilha and Thiago V. Spina},
title={cudaIFT: 180x Faster Image Foresting Transform for Waterpixel Estimation using CUDA},
booktitle={Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP},
year={2019},
pages={395-404},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007402703950404},
isbn={978-989-758-354-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2019) - Volume 4: VISAPP
TI - cudaIFT: 180x Faster Image Foresting Transform for Waterpixel Estimation using CUDA
SN - 978-989-758-354-4
AU - Gonçalves H.
AU - Q. de Vasconcelos G.
AU - Rangel P.
AU - Carvalho M.
AU - Archilha N.
AU - Spina T.
PY - 2019
SP - 395
EP - 404
DO - 10.5220/0007402703950404
PB - SciTePress