Dense motion estimation for smoke

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Motion estimation for highly dynamic phenomena such as smoke is an open challenge for Computer Vision. Traditional dense motion estimation algorithms have difficulties with non-rigid and large motions, both of which are frequently observed in smoke motion. We propose an algorithm for dense motion estimation of smoke. Our algorithm is robust, fast, and has better performance over different types of smoke compared to other dense motion estimation algorithms, including state of the art and neural network approaches. The key to our contribution is to use skeletal flow, without explicit point matching, to provide a sparse flow. This sparse flow is upgraded to a dense flow. In this paper we describe our algorithm in greater detail, and provide experimental evidence to support our claims.

LanguageEnglish
Title of host publicationComputer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers
EditorsK. Nishino, S-H. Lai, V. Lepetit, Y. Sato
PublisherSpringer Nature
Pages225-239
Number of pages15
ISBN (Electronic)978-3-319-54190-7
ISBN (Print)9783319541891
DOIs
StatusPublished - 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10114

Fingerprint

Motion estimation
Smoke
Computer vision
Neural networks

Cite this

Chen, D., Li, W., & Hall, P. (2017). Dense motion estimation for smoke. In K. Nishino, S-H. Lai, V. Lepetit, & Y. Sato (Eds.), Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers (pp. 225-239). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10114 ). Springer Nature. DOI: 10.1007/978-3-319-54190-7_14

Dense motion estimation for smoke. / Chen, Da; Li, Wenbin; Hall, Peter.

Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. ed. / K. Nishino; S-H. Lai; V. Lepetit; Y. Sato. Springer Nature, 2017. p. 225-239 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10114 ).

Research output: Chapter in Book/Report/Conference proceedingChapter

Chen, D, Li, W & Hall, P 2017, Dense motion estimation for smoke. in K Nishino, S-H Lai, V Lepetit & Y Sato (eds), Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10114 , Springer Nature, pp. 225-239. DOI: 10.1007/978-3-319-54190-7_14
Chen D, Li W, Hall P. Dense motion estimation for smoke. In Nishino K, Lai S-H, Lepetit V, Sato Y, editors, Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. Springer Nature. 2017. p. 225-239. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Available from, DOI: 10.1007/978-3-319-54190-7_14
Chen, Da ; Li, Wenbin ; Hall, Peter. / Dense motion estimation for smoke. Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. editor / K. Nishino ; S-H. Lai ; V. Lepetit ; Y. Sato. Springer Nature, 2017. pp. 225-239 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inbook{3b2d4886177849359a7761adf99596f3,
title = "Dense motion estimation for smoke",
abstract = "Motion estimation for highly dynamic phenomena such as smoke is an open challenge for Computer Vision. Traditional dense motion estimation algorithms have difficulties with non-rigid and large motions, both of which are frequently observed in smoke motion. We propose an algorithm for dense motion estimation of smoke. Our algorithm is robust, fast, and has better performance over different types of smoke compared to other dense motion estimation algorithms, including state of the art and neural network approaches. The key to our contribution is to use skeletal flow, without explicit point matching, to provide a sparse flow. This sparse flow is upgraded to a dense flow. In this paper we describe our algorithm in greater detail, and provide experimental evidence to support our claims.",
author = "Da Chen and Wenbin Li and Peter Hall",
year = "2017",
doi = "10.1007/978-3-319-54190-7_14",
language = "English",
isbn = "9783319541891",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Nature",
pages = "225--239",
editor = "K. Nishino and S-H. Lai and V. Lepetit and Y. Sato",
booktitle = "Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers",
address = "USA United States",

}

TY - CHAP

T1 - Dense motion estimation for smoke

AU - Chen,Da

AU - Li,Wenbin

AU - Hall,Peter

PY - 2017

Y1 - 2017

N2 - Motion estimation for highly dynamic phenomena such as smoke is an open challenge for Computer Vision. Traditional dense motion estimation algorithms have difficulties with non-rigid and large motions, both of which are frequently observed in smoke motion. We propose an algorithm for dense motion estimation of smoke. Our algorithm is robust, fast, and has better performance over different types of smoke compared to other dense motion estimation algorithms, including state of the art and neural network approaches. The key to our contribution is to use skeletal flow, without explicit point matching, to provide a sparse flow. This sparse flow is upgraded to a dense flow. In this paper we describe our algorithm in greater detail, and provide experimental evidence to support our claims.

AB - Motion estimation for highly dynamic phenomena such as smoke is an open challenge for Computer Vision. Traditional dense motion estimation algorithms have difficulties with non-rigid and large motions, both of which are frequently observed in smoke motion. We propose an algorithm for dense motion estimation of smoke. Our algorithm is robust, fast, and has better performance over different types of smoke compared to other dense motion estimation algorithms, including state of the art and neural network approaches. The key to our contribution is to use skeletal flow, without explicit point matching, to provide a sparse flow. This sparse flow is upgraded to a dense flow. In this paper we describe our algorithm in greater detail, and provide experimental evidence to support our claims.

UR - http://www.scopus.com/inward/record.url?scp=85016029858&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1007/978-3-319-54190-7_14

U2 - 10.1007/978-3-319-54190-7_14

DO - 10.1007/978-3-319-54190-7_14

M3 - Chapter

SN - 9783319541891

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 225

EP - 239

BT - Computer Vision -ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers

PB - Springer Nature

ER -