Estimation of motion and structure of planar surfaces from a sequence of monocular images

Sanghoon Sull, Narendra Ahuja

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

An algorithm is presented which estimates 10 parameters for motion and structure of a rigid planar patch given point correspondences in a monocular image sequence under perspective projection. The rotational velocity is assumed constant and the rotation center arbitrary. The algorithm mainly consists of two steps. First, the 3-D space of (ωx, ωy, ωz) is searched exhaustively and for each (ωx, ωy, ωz) all the other parameters with the value of an objective function are computed linearly. Some of the (ωx, ωy, ωz) and the corresponding structure values are used as the initial guesses in the second step. The objective function is iteratively minimized with respect to five variables for rotation and structure. The solution corresponding to the global minimum is used to obtain least squares estimates of the remaining unknowns, for translation and rotation center. It was experimentally found that the objective function converges well so that the 3-D space need not be densely searched. Results are presented for three image sequences, two simulated and one real.

Original languageEnglish
Title of host publicationUnknown Host Publication Title
Place of PublicationPiscataway, NJ, United States
PublisherPubl by IEEE
Pages732-733
Number of pages2
ISBN (Print)0818621486
Publication statusPublished - 1991 Dec 1
EventProceedings of the 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Maui, HI
Duration: 1991 Jun 31991 Jun 6

Other

OtherProceedings of the 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CityMaui, HI
Period91/6/391/6/6

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Sull, S., & Ahuja, N. (1991). Estimation of motion and structure of planar surfaces from a sequence of monocular images. In Unknown Host Publication Title (pp. 732-733). Piscataway, NJ, United States: Publ by IEEE.

Estimation of motion and structure of planar surfaces from a sequence of monocular images. / Sull, Sanghoon; Ahuja, Narendra.

Unknown Host Publication Title. Piscataway, NJ, United States : Publ by IEEE, 1991. p. 732-733.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sull, S & Ahuja, N 1991, Estimation of motion and structure of planar surfaces from a sequence of monocular images. in Unknown Host Publication Title. Publ by IEEE, Piscataway, NJ, United States, pp. 732-733, Proceedings of the 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Maui, HI, 91/6/3.
Sull S, Ahuja N. Estimation of motion and structure of planar surfaces from a sequence of monocular images. In Unknown Host Publication Title. Piscataway, NJ, United States: Publ by IEEE. 1991. p. 732-733
Sull, Sanghoon ; Ahuja, Narendra. / Estimation of motion and structure of planar surfaces from a sequence of monocular images. Unknown Host Publication Title. Piscataway, NJ, United States : Publ by IEEE, 1991. pp. 732-733
@inproceedings{033b0ebe649a4eda89a5514069689af2,
title = "Estimation of motion and structure of planar surfaces from a sequence of monocular images",
abstract = "An algorithm is presented which estimates 10 parameters for motion and structure of a rigid planar patch given point correspondences in a monocular image sequence under perspective projection. The rotational velocity is assumed constant and the rotation center arbitrary. The algorithm mainly consists of two steps. First, the 3-D space of (ωx, ωy, ωz) is searched exhaustively and for each (ωx, ωy, ωz) all the other parameters with the value of an objective function are computed linearly. Some of the (ωx, ωy, ωz) and the corresponding structure values are used as the initial guesses in the second step. The objective function is iteratively minimized with respect to five variables for rotation and structure. The solution corresponding to the global minimum is used to obtain least squares estimates of the remaining unknowns, for translation and rotation center. It was experimentally found that the objective function converges well so that the 3-D space need not be densely searched. Results are presented for three image sequences, two simulated and one real.",
author = "Sanghoon Sull and Narendra Ahuja",
year = "1991",
month = "12",
day = "1",
language = "English",
isbn = "0818621486",
pages = "732--733",
booktitle = "Unknown Host Publication Title",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Estimation of motion and structure of planar surfaces from a sequence of monocular images

AU - Sull, Sanghoon

AU - Ahuja, Narendra

PY - 1991/12/1

Y1 - 1991/12/1

N2 - An algorithm is presented which estimates 10 parameters for motion and structure of a rigid planar patch given point correspondences in a monocular image sequence under perspective projection. The rotational velocity is assumed constant and the rotation center arbitrary. The algorithm mainly consists of two steps. First, the 3-D space of (ωx, ωy, ωz) is searched exhaustively and for each (ωx, ωy, ωz) all the other parameters with the value of an objective function are computed linearly. Some of the (ωx, ωy, ωz) and the corresponding structure values are used as the initial guesses in the second step. The objective function is iteratively minimized with respect to five variables for rotation and structure. The solution corresponding to the global minimum is used to obtain least squares estimates of the remaining unknowns, for translation and rotation center. It was experimentally found that the objective function converges well so that the 3-D space need not be densely searched. Results are presented for three image sequences, two simulated and one real.

AB - An algorithm is presented which estimates 10 parameters for motion and structure of a rigid planar patch given point correspondences in a monocular image sequence under perspective projection. The rotational velocity is assumed constant and the rotation center arbitrary. The algorithm mainly consists of two steps. First, the 3-D space of (ωx, ωy, ωz) is searched exhaustively and for each (ωx, ωy, ωz) all the other parameters with the value of an objective function are computed linearly. Some of the (ωx, ωy, ωz) and the corresponding structure values are used as the initial guesses in the second step. The objective function is iteratively minimized with respect to five variables for rotation and structure. The solution corresponding to the global minimum is used to obtain least squares estimates of the remaining unknowns, for translation and rotation center. It was experimentally found that the objective function converges well so that the 3-D space need not be densely searched. Results are presented for three image sequences, two simulated and one real.

UR - http://www.scopus.com/inward/record.url?scp=0026390824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026390824&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026390824

SN - 0818621486

SP - 732

EP - 733

BT - Unknown Host Publication Title

PB - Publ by IEEE

CY - Piscataway, NJ, United States

ER -