Superpixel hierarchy

Xing Wei, Qingxiong Yang, Yihong Gong, Narendra Ahuja, Ming Hsuan Yang

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Superpixel segmentation has been one of the most important tasks in computer vision. In practice, an object can be represented by a number of segments at finer levels with consistent details or included in a surrounding region at coarser levels. Thus, a superpixel segmentation hierarchy is of great importance for applications that require different levels of image details. However, there is no method that can generate all scales of superpixels accurately in real time. In this paper, we propose the superhierarchy algorithm which is able to generate multi-scale superpixels as accurately as the state-of-the-art methods but with one to two orders of magnitude speed-up. The proposed algorithm can be directly integrated with recent efficient edge detectors to significantly outperform the state-of-the-art methods in terms of segmentation accuracy. Quantitative and qualitative evaluations on a number of applications demonstrate that the proposed algorithm is accurate and efficient in generating a hierarchy of superpixels.

Original languageEnglish
Pages (from-to)4838-4849
Number of pages12
JournalIEEE Transactions on Image Processing
Volume27
Issue number10
DOIs
Publication statusPublished - 2018 Oct

Fingerprint

Computer vision
Detectors

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Wei, Xing ; Yang, Qingxiong ; Gong, Yihong ; Ahuja, Narendra ; Yang, Ming Hsuan. / Superpixel hierarchy. In: IEEE Transactions on Image Processing. 2018 ; Vol. 27, No. 10. pp. 4838-4849.
@article{8f343ed58e9d46a8813952a9ac30fc7a,
title = "Superpixel hierarchy",
abstract = "Superpixel segmentation has been one of the most important tasks in computer vision. In practice, an object can be represented by a number of segments at finer levels with consistent details or included in a surrounding region at coarser levels. Thus, a superpixel segmentation hierarchy is of great importance for applications that require different levels of image details. However, there is no method that can generate all scales of superpixels accurately in real time. In this paper, we propose the superhierarchy algorithm which is able to generate multi-scale superpixels as accurately as the state-of-the-art methods but with one to two orders of magnitude speed-up. The proposed algorithm can be directly integrated with recent efficient edge detectors to significantly outperform the state-of-the-art methods in terms of segmentation accuracy. Quantitative and qualitative evaluations on a number of applications demonstrate that the proposed algorithm is accurate and efficient in generating a hierarchy of superpixels.",
author = "Xing Wei and Qingxiong Yang and Yihong Gong and Narendra Ahuja and Yang, {Ming Hsuan}",
year = "2018",
month = "10",
doi = "10.1109/TIP.2018.2836300",
language = "English",
volume = "27",
pages = "4838--4849",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "10",

}

Wei, X, Yang, Q, Gong, Y, Ahuja, N & Yang, MH 2018, 'Superpixel hierarchy', IEEE Transactions on Image Processing, vol. 27, no. 10, pp. 4838-4849. https://doi.org/10.1109/TIP.2018.2836300

Superpixel hierarchy. / Wei, Xing; Yang, Qingxiong; Gong, Yihong; Ahuja, Narendra; Yang, Ming Hsuan.

In: IEEE Transactions on Image Processing, Vol. 27, No. 10, 10.2018, p. 4838-4849.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Superpixel hierarchy

AU - Wei, Xing

AU - Yang, Qingxiong

AU - Gong, Yihong

AU - Ahuja, Narendra

AU - Yang, Ming Hsuan

PY - 2018/10

Y1 - 2018/10

N2 - Superpixel segmentation has been one of the most important tasks in computer vision. In practice, an object can be represented by a number of segments at finer levels with consistent details or included in a surrounding region at coarser levels. Thus, a superpixel segmentation hierarchy is of great importance for applications that require different levels of image details. However, there is no method that can generate all scales of superpixels accurately in real time. In this paper, we propose the superhierarchy algorithm which is able to generate multi-scale superpixels as accurately as the state-of-the-art methods but with one to two orders of magnitude speed-up. The proposed algorithm can be directly integrated with recent efficient edge detectors to significantly outperform the state-of-the-art methods in terms of segmentation accuracy. Quantitative and qualitative evaluations on a number of applications demonstrate that the proposed algorithm is accurate and efficient in generating a hierarchy of superpixels.

AB - Superpixel segmentation has been one of the most important tasks in computer vision. In practice, an object can be represented by a number of segments at finer levels with consistent details or included in a surrounding region at coarser levels. Thus, a superpixel segmentation hierarchy is of great importance for applications that require different levels of image details. However, there is no method that can generate all scales of superpixels accurately in real time. In this paper, we propose the superhierarchy algorithm which is able to generate multi-scale superpixels as accurately as the state-of-the-art methods but with one to two orders of magnitude speed-up. The proposed algorithm can be directly integrated with recent efficient edge detectors to significantly outperform the state-of-the-art methods in terms of segmentation accuracy. Quantitative and qualitative evaluations on a number of applications demonstrate that the proposed algorithm is accurate and efficient in generating a hierarchy of superpixels.

UR - http://www.scopus.com/inward/record.url?scp=85046998367&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046998367&partnerID=8YFLogxK

U2 - 10.1109/TIP.2018.2836300

DO - 10.1109/TIP.2018.2836300

M3 - Article

AN - SCOPUS:85046998367

VL - 27

SP - 4838

EP - 4849

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 10

ER -