Fast Superpixel Segmentation via Boundary Sampling and Interpolation

Li XU; Bing LUO; Mingming KONG; Bo LI; Zheng PEI

doi:10.1587/transinf.2018EDL8168

Fast Superpixel Segmentation via Boundary Sampling and Interpolation

Li XU, Bing LUO, Mingming KONG, Bo LI, Zheng PEI

Full Text Views

0

Cite this

Summary :

This letter proposes a fast superpixel segmentation method based on boundary sampling and interpolation. The basic idea is as follow: instead of labeling local region pixels, we estimate superpixel boundary by interpolating candidate boundary pixel from a down-sampling image segmentation. On the one hand, there exists high spatial redundancy within each local region, which could be discarded. On the other hand, we estimate the labels of candidate boundary pixels via sampling superpixel boundary within corresponding neighbour. Benefiting from the reduction of candidate pixel distance calculation, the proposed method significantly accelerates superpixel segmentation. Experiments on BSD500 benchmark demonstrate that our method needs half the time compared with the state-of-the-arts while almost no accuracy reduction.

Publication: IEICE TRANSACTIONS on Information Vol.E102-D No.4 pp.871-874

Publication Date: 2019/04/01

Publicized: 2019/01/22

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2018EDL8168

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Authors

Li XU
  Xihua University
Bing LUO
  Xihua University
Mingming KONG
  Xihua University
Bo LI
  Southwest Jiaotong University
Zheng PEI
  Xihua University

Keyword

superpixel, acceleration, sampling, interpolation

Cite this

Copy

Li XU, Bing LUO, Mingming KONG, Bo LI, Zheng PEI, "Fast Superpixel Segmentation via Boundary Sampling and Interpolation" in IEICE TRANSACTIONS on Information, vol. E102-D, no. 4, pp. 871-874, April 2019, doi: 10.1587/transinf.2018EDL8168.
Abstract: This letter proposes a fast superpixel segmentation method based on boundary sampling and interpolation. The basic idea is as follow: instead of labeling local region pixels, we estimate superpixel boundary by interpolating candidate boundary pixel from a down-sampling image segmentation. On the one hand, there exists high spatial redundancy within each local region, which could be discarded. On the other hand, we estimate the labels of candidate boundary pixels via sampling superpixel boundary within corresponding neighbour. Benefiting from the reduction of candidate pixel distance calculation, the proposed method significantly accelerates superpixel segmentation. Experiments on BSD500 benchmark demonstrate that our method needs half the time compared with the state-of-the-arts while almost no accuracy reduction.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8168/_p

Copy

@ARTICLE{e102-d_4_871,
author={Li XU, Bing LUO, Mingming KONG, Bo LI, Zheng PEI, },
journal={IEICE TRANSACTIONS on Information},
title={Fast Superpixel Segmentation via Boundary Sampling and Interpolation},
year={2019},
volume={E102-D},
number={4},
pages={871-874},
abstract={This letter proposes a fast superpixel segmentation method based on boundary sampling and interpolation. The basic idea is as follow: instead of labeling local region pixels, we estimate superpixel boundary by interpolating candidate boundary pixel from a down-sampling image segmentation. On the one hand, there exists high spatial redundancy within each local region, which could be discarded. On the other hand, we estimate the labels of candidate boundary pixels via sampling superpixel boundary within corresponding neighbour. Benefiting from the reduction of candidate pixel distance calculation, the proposed method significantly accelerates superpixel segmentation. Experiments on BSD500 benchmark demonstrate that our method needs half the time compared with the state-of-the-arts while almost no accuracy reduction.},
keywords={},
doi={10.1587/transinf.2018EDL8168},
ISSN={1745-1361},
month={April},}

Copy

TY - JOUR
TI - Fast Superpixel Segmentation via Boundary Sampling and Interpolation
T2 - IEICE TRANSACTIONS on Information
SP - 871
EP - 874
AU - Li XU
AU - Bing LUO
AU - Mingming KONG
AU - Bo LI
AU - Zheng PEI
PY - 2019
DO - 10.1587/transinf.2018EDL8168
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2019
AB - This letter proposes a fast superpixel segmentation method based on boundary sampling and interpolation. The basic idea is as follow: instead of labeling local region pixels, we estimate superpixel boundary by interpolating candidate boundary pixel from a down-sampling image segmentation. On the one hand, there exists high spatial redundancy within each local region, which could be discarded. On the other hand, we estimate the labels of candidate boundary pixels via sampling superpixel boundary within corresponding neighbour. Benefiting from the reduction of candidate pixel distance calculation, the proposed method significantly accelerates superpixel segmentation. Experiments on BSD500 benchmark demonstrate that our method needs half the time compared with the state-of-the-arts while almost no accuracy reduction.
ER -