Fitting quadrics with a Bayesian prior

Daniel Beale; Peter Hall; Darren Cosker; Yongliang Yang; Neill Campbell

doi:10.1007/s41095-016-0041-9

Fitting quadrics with a Bayesian prior

Daniel Beale, Peter Hall, Darren Cosker, Yongliang Yang, Neill Campbell

Research output: Contribution to conference › Paper › peer-review

Abstract

Quadrics are a compact mathematical formulation for a range of primitive surfaces. A problem arises when there are not enough data-points to compute the model but knowledge of the shape is available. This paper presents a method for fitting a quadric with a Bayesian prior. We use a matrix normal prior in order to favour ellipsoids on ambiguous data. The results show the algorithm to cope well when there are few points in the point cloud, competing with contemporary techniques in the area.

Original language	English
DOIs	https://doi.org/10.1007/s41095-016-0041-9
Publication status	Published - Jun 2016
Event	4th International conference on Computational Visual Media (CVM 2016) - Wales, Cardiff, UK United Kingdom Duration: 6 Apr 2016 → 8 Apr 2016

Conference

Conference	4th International conference on Computational Visual Media (CVM 2016)
Country/Territory	UK United Kingdom
City	Cardiff
Period	6/04/16 → 8/04/16

Access to Document

10.1007/s41095-016-0041-9

Cite this

@conference{74c03a7db9394c3eb9f4869a2083140e,

title = "Fitting quadrics with a Bayesian prior",

abstract = "Quadrics are a compact mathematical formulation for a range of primitive surfaces. A problem arises when there are not enough data-points to compute the model but knowledge of the shape is available. This paper presents a method for fitting a quadric with a Bayesian prior. We use a matrix normal prior in order to favour ellipsoids on ambiguous data. The results show the algorithm to cope well when there are few points in the point cloud, competing with contemporary techniques in the area.",

author = "Daniel Beale and Peter Hall and Darren Cosker and Yongliang Yang and Neill Campbell",

year = "2016",

month = jun,

doi = "10.1007/s41095-016-0041-9",

language = "English",

note = "4th International conference on Computational Visual Media (CVM 2016) ; Conference date: 06-04-2016 Through 08-04-2016",

}

TY - CONF

T1 - Fitting quadrics with a Bayesian prior

AU - Beale, Daniel

AU - Hall, Peter

AU - Cosker, Darren

AU - Yang, Yongliang

AU - Campbell, Neill

PY - 2016/6

Y1 - 2016/6

N2 - Quadrics are a compact mathematical formulation for a range of primitive surfaces. A problem arises when there are not enough data-points to compute the model but knowledge of the shape is available. This paper presents a method for fitting a quadric with a Bayesian prior. We use a matrix normal prior in order to favour ellipsoids on ambiguous data. The results show the algorithm to cope well when there are few points in the point cloud, competing with contemporary techniques in the area.

AB - Quadrics are a compact mathematical formulation for a range of primitive surfaces. A problem arises when there are not enough data-points to compute the model but knowledge of the shape is available. This paper presents a method for fitting a quadric with a Bayesian prior. We use a matrix normal prior in order to favour ellipsoids on ambiguous data. The results show the algorithm to cope well when there are few points in the point cloud, competing with contemporary techniques in the area.

UR - http://dx.doi.org/10.1007/s41095-016-0041-9

U2 - 10.1007/s41095-016-0041-9

DO - 10.1007/s41095-016-0041-9

M3 - Paper

T2 - 4th International conference on Computational Visual Media (CVM 2016)

Y2 - 6 April 2016 through 8 April 2016

ER -

Fitting quadrics with a Bayesian prior

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Centre for the Analysis of Motion, Entertainment Research and Applications (CAMERA)

Cite this

Fitting quadrics with a Bayesian prior

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Projects

Centre for the Analysis of Motion, Entertainment Research and Applications (CAMERA)

Cite this