Stereo superpixel : An iterative framework based on parallax consistency and collaborative optimization

Hua LI; Runmin CONG; Sam KWONG; Chuanbo CHEN; Qianqian XU; Chongyi LI

doi:10.1016/j.ins.2020.12.031

Stereo superpixel : An iterative framework based on parallax consistency and collaborative optimization

Hua LI, Runmin CONG, Sam KWONG, Chuanbo CHEN, Qianqian XU, Chongyi LI

Research output: Journal Publications › Journal Article (refereed) › peer-review

9 Citations (Scopus)

Abstract

Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation.

Original language	English
Pages (from-to)	209-222
Journal	Information Sciences
Volume	556
Early online date	18 Jan 2021
DOIs	https://doi.org/10.1016/j.ins.2020.12.031
Publication status	Published - May 2021
Externally published	Yes

Funding

This work was supported by the Key Project of Science and Technology Innovation 2030 supported by the Ministry of Science and Technology of China under Grant 2018AAA0101301, in part by the Natural Science Foundation of China under Grants 61772344, 62002014, in part by the Hong Kong RGC General Research Funds under 9042816 (CityU 11209819), in part by the Beijing Nova Program under Grant Z201100006820016, in part by the Fundamental Research Funds for the Central Universities under Grant 2019RC039, in part by Elite Scientist Sponsorship Program by the Beijing Association for Science and Technology, in part by Hong Kong Scholars Program, and in part by China Postdoctoral Science Foundation under Grant 2020T130050, Grant 2019M660438.

Keywords

Collaborative optimization
Parallax consistency
Stereo superpixel
Superpixel segmentation

Access to Document

10.1016/j.ins.2020.12.031

Cite this

@article{55e361f2098a4379b2354248a9146f96,

title = "Stereo superpixel : An iterative framework based on parallax consistency and collaborative optimization",

abstract = "Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation.",

keywords = "Collaborative optimization, Parallax consistency, Stereo superpixel, Superpixel segmentation",

author = "Hua LI and Runmin CONG and Sam KWONG and Chuanbo CHEN and Qianqian XU and Chongyi LI",

year = "2021",

month = may,

doi = "10.1016/j.ins.2020.12.031",

language = "English",

volume = "556",

pages = "209--222",

journal = "Information Sciences",

issn = "0020-0255",

publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Stereo superpixel : An iterative framework based on parallax consistency and collaborative optimization

AU - LI, Hua

AU - CONG, Runmin

AU - KWONG, Sam

AU - CHEN, Chuanbo

AU - XU, Qianqian

AU - LI, Chongyi

PY - 2021/5

Y1 - 2021/5

N2 - Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation.

AB - Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation.

KW - Collaborative optimization

KW - Parallax consistency

KW - Stereo superpixel

KW - Superpixel segmentation

UR - http://www.scopus.com/inward/record.url?scp=85100166036&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2020.12.031

DO - 10.1016/j.ins.2020.12.031

M3 - Journal Article (refereed)

SN - 0020-0255

VL - 556

SP - 209

EP - 222

JO - Information Sciences

JF - Information Sciences

ER -

Stereo superpixel : An iterative framework based on parallax consistency and collaborative optimization

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this