SDICP: Semi-Dense Tracking based on Iterative Closest Points

Laurent Kneip; Yi Zhou; Hongdong Li

doi:10.5244/C.29.100

SDICP: Semi-Dense Tracking based on Iterative Closest Points

Laurent Kneip, Yi Zhou, Hongdong Li

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper introduces a novel strategy for real-time monocular camera tracking over the recently introduced, efficient semi-dense depth maps. We employ a geometric iterative closest point technique instead of a photometric error criterion, which has the conceptual advantage of requiring neither isotropic enlargement of the employed semidense regions, nor pyramidal subsampling. We outline the detailed concepts leading to robustness and efficiency even for large frame-to-frame disparities. We demonstrate successful real-time processing over very large view-point changes and significantly corrupted semi-dense depth-maps, thus underlining the validity of our geometric approach

Original language	English
Title of host publication	Proceedings of the British Machine Vision Conference 2015
Editors	Xianghua Xie, Mark W. Jones, and Gary K. L. Tam
Place of Publication	Swansea
Publisher	British Machine Vision Association, BMVA
Pages	1-12
Edition	Peer Reviewed
ISBN (Print)	9781901725537
DOIs	https://doi.org/10.5244/C.29.100
Publication status	Published - 2015
Event	British Machine Vision Conference BMVC 2015 - Swansea, UK Duration: 1 Jan 2015 → … http://www.bmva.org/bmvc/2015/index.html

Conference

Conference	British Machine Vision Conference BMVC 2015
Period	1/01/15 → …
Other	September 7-10 2015
Internet address	http://www.bmva.org/bmvc/2015/index.html

Access to Document

10.5244/C.29.100

Cite this

@inproceedings{0c5a7ab9646e4db68279bc15b75c88f4,

title = "SDICP: Semi-Dense Tracking based on Iterative Closest Points",

abstract = "This paper introduces a novel strategy for real-time monocular camera tracking over the recently introduced, efficient semi-dense depth maps. We employ a geometric iterative closest point technique instead of a photometric error criterion, which has the conceptual advantage of requiring neither isotropic enlargement of the employed semidense regions, nor pyramidal subsampling. We outline the detailed concepts leading to robustness and efficiency even for large frame-to-frame disparities. We demonstrate successful real-time processing over very large view-point changes and significantly corrupted semi-dense depth-maps, thus underlining the validity of our geometric approach",

author = "Laurent Kneip and Yi Zhou and Hongdong Li",

year = "2015",

doi = "10.5244/C.29.100",

language = "English",

isbn = "9781901725537",

pages = "1--12",

editor = "{Xianghua Xie, Mark W. Jones, and Gary K. L. Tam}",

booktitle = "Proceedings of the British Machine Vision Conference 2015",

publisher = "British Machine Vision Association, BMVA",

edition = "Peer Reviewed",

note = "British Machine Vision Conference BMVC 2015 ; Conference date: 01-01-2015",

url = "http://www.bmva.org/bmvc/2015/index.html",

}

SDICP: Semi-Dense Tracking based on Iterative Closest Points. / Kneip, Laurent; Zhou, Yi; Li, Hongdong.
Proceedings of the British Machine Vision Conference 2015. ed. / Xianghua Xie, Mark W. Jones, and Gary K. L. Tam. Peer Reviewed. ed. Swansea: British Machine Vision Association, BMVA, 2015. p. 1-12.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - SDICP: Semi-Dense Tracking based on Iterative Closest Points

AU - Kneip, Laurent

AU - Zhou, Yi

AU - Li, Hongdong

PY - 2015

Y1 - 2015

N2 - This paper introduces a novel strategy for real-time monocular camera tracking over the recently introduced, efficient semi-dense depth maps. We employ a geometric iterative closest point technique instead of a photometric error criterion, which has the conceptual advantage of requiring neither isotropic enlargement of the employed semidense regions, nor pyramidal subsampling. We outline the detailed concepts leading to robustness and efficiency even for large frame-to-frame disparities. We demonstrate successful real-time processing over very large view-point changes and significantly corrupted semi-dense depth-maps, thus underlining the validity of our geometric approach

AB - This paper introduces a novel strategy for real-time monocular camera tracking over the recently introduced, efficient semi-dense depth maps. We employ a geometric iterative closest point technique instead of a photometric error criterion, which has the conceptual advantage of requiring neither isotropic enlargement of the employed semidense regions, nor pyramidal subsampling. We outline the detailed concepts leading to robustness and efficiency even for large frame-to-frame disparities. We demonstrate successful real-time processing over very large view-point changes and significantly corrupted semi-dense depth-maps, thus underlining the validity of our geometric approach

U2 - 10.5244/C.29.100

DO - 10.5244/C.29.100

M3 - Conference contribution

SN - 9781901725537

SP - 1

EP - 12

BT - Proceedings of the British Machine Vision Conference 2015

A2 - null, Xianghua Xie, Mark W. Jones, and Gary K. L. Tam

PB - British Machine Vision Association, BMVA

CY - Swansea

T2 - British Machine Vision Conference BMVC 2015

Y2 - 1 January 2015

ER -

SDICP: Semi-Dense Tracking based on Iterative Closest Points

Abstract

Conference

Access to Document

Fingerprint

Cite this