3D Model Acquisition by Tracking 2DWireframes

Matthew Brown, T Drummond, R Cipolla

Research output: Contribution to conferencePaper

Abstract

This paper presents a semi-automatic wireframe acquisition system. The system uses real-time (25Hz) tracking of a user specified 2D wireframe and intermittent camera pose parameters to accumulate 3D position information. The 2D tracking framework enables the application of model-based constraints in an intuitive way which interacts naturally with the Kalman filter formulation used. In particular, this is used to introduce feedback from the current 3D shape estimate to improve the robustness of the 2D tracking. The scheme allows wireframe models of simple edge based objects to be built in around 5 minutes.
Original languageEnglish
Pages656-665
Number of pages10
Publication statusPublished - Sep 2000
EventBMVC 200: Eleventh British Machine Vision Conference - Bristol
Duration: 11 Sep 200014 Sep 2000

Conference

ConferenceBMVC 200: Eleventh British Machine Vision Conference
CityBristol
Period11/09/0014/09/00

Fingerprint

Feedback

Cite this

Brown, M., Drummond, T., & Cipolla, R. (2000). 3D Model Acquisition by Tracking 2DWireframes. 656-665. Paper presented at BMVC 200: Eleventh British Machine Vision Conference, Bristol, .

3D Model Acquisition by Tracking 2DWireframes. / Brown, Matthew; Drummond, T; Cipolla, R.

2000. 656-665 Paper presented at BMVC 200: Eleventh British Machine Vision Conference, Bristol, .

Research output: Contribution to conferencePaper

Brown, M, Drummond, T & Cipolla, R 2000, '3D Model Acquisition by Tracking 2DWireframes' Paper presented at BMVC 200: Eleventh British Machine Vision Conference, Bristol, 11/09/00 - 14/09/00, pp. 656-665.
Brown M, Drummond T, Cipolla R. 3D Model Acquisition by Tracking 2DWireframes. 2000. Paper presented at BMVC 200: Eleventh British Machine Vision Conference, Bristol, .
Brown, Matthew ; Drummond, T ; Cipolla, R. / 3D Model Acquisition by Tracking 2DWireframes. Paper presented at BMVC 200: Eleventh British Machine Vision Conference, Bristol, .10 p.
@conference{2eb04ffe19304464be3f472455c55014,
title = "3D Model Acquisition by Tracking 2DWireframes",
abstract = "This paper presents a semi-automatic wireframe acquisition system. The system uses real-time (25Hz) tracking of a user specified 2D wireframe and intermittent camera pose parameters to accumulate 3D position information. The 2D tracking framework enables the application of model-based constraints in an intuitive way which interacts naturally with the Kalman filter formulation used. In particular, this is used to introduce feedback from the current 3D shape estimate to improve the robustness of the 2D tracking. The scheme allows wireframe models of simple edge based objects to be built in around 5 minutes.",
author = "Matthew Brown and T Drummond and R Cipolla",
year = "2000",
month = "9",
language = "English",
pages = "656--665",
note = "BMVC 200: Eleventh British Machine Vision Conference ; Conference date: 11-09-2000 Through 14-09-2000",

}

TY - CONF

T1 - 3D Model Acquisition by Tracking 2DWireframes

AU - Brown, Matthew

AU - Drummond, T

AU - Cipolla, R

PY - 2000/9

Y1 - 2000/9

N2 - This paper presents a semi-automatic wireframe acquisition system. The system uses real-time (25Hz) tracking of a user specified 2D wireframe and intermittent camera pose parameters to accumulate 3D position information. The 2D tracking framework enables the application of model-based constraints in an intuitive way which interacts naturally with the Kalman filter formulation used. In particular, this is used to introduce feedback from the current 3D shape estimate to improve the robustness of the 2D tracking. The scheme allows wireframe models of simple edge based objects to be built in around 5 minutes.

AB - This paper presents a semi-automatic wireframe acquisition system. The system uses real-time (25Hz) tracking of a user specified 2D wireframe and intermittent camera pose parameters to accumulate 3D position information. The 2D tracking framework enables the application of model-based constraints in an intuitive way which interacts naturally with the Kalman filter formulation used. In particular, this is used to introduce feedback from the current 3D shape estimate to improve the robustness of the 2D tracking. The scheme allows wireframe models of simple edge based objects to be built in around 5 minutes.

UR - http://www.bmva.org/bmvc/2000/papers/p66.pdf

UR - http://www.bmva.org/bmvc/2000/index.html

M3 - Paper

SP - 656

EP - 665

ER -