University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Bayesian Helmholtz Stereopsis with Integrability Prior

Roubtsova, Nadejda and Guillemaut, Jean-Yves (2017) Bayesian Helmholtz Stereopsis with Integrability Prior IEEE Transactions on Pattern Analysis and Machine Intelligence, 40 (9). pp. 2265-2272.

[img]
Preview
Text
Bayesian Helmholtz Stereopsis with Integrability Prior.pdf - Version of Record
Available under License Creative Commons Attribution.

Download (860kB) | Preview

Abstract

Helmholtz Stereopsis is a 3D reconstruction method uniquely independent of surface reflectance. Yet, its sub-optimal maximum likelihood formulation with drift-prone normal integration limits performance. Via three contributions this paper presents a complete novel pipeline for Helmholtz Stereopsis. Firstly, we propose a Bayesian formulation replacing the maximum likelihood problem by a maximum a posteriori one. Secondly, a tailored prior enforcing consistency between depth and normal estimates via a novel metric related to optimal surface integrability is proposed. Thirdly, explicit surface integration is eliminated by taking advantage of the accuracy of prior and high resolution of the coarse-to-fine approach. The pipeline is validated quantitatively and qualitatively against alternative formulations, reaching sub-millimetre accuracy and coping with complex geometry and reflectance.

Item Type: Article
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :
NameEmailORCID
Roubtsova, Nadejdan.s.roubtsova@surrey.ac.uk
Guillemaut, Jean-YvesJ.Guillemaut@surrey.ac.uk
Date : 22 September 2017
Funders : Engineering and Physical Sciences Research Council (EPSRC)
DOI : 10.1109/TPAMI.2017.2749373
Copyright Disclaimer : This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/
Uncontrolled Keywords : Helmholtz Stereopsis; 3D; Complex reflectance; MAP
Related URLs :
Additional Information : Our synthetic data is available at https://doi.org/10.15126/surreydata.00841369. Source of 3D models used (pear/bunny): Suggestive Contour model database, http://gfx.cs.princeton.edu/proj/sugcon/models/ and Stanford 3D scanning repository, https://graphics.stanford.edu/data/3Dscanrep/ respectively.
Depositing User : Clive Harris
Date Deposited : 05 Sep 2017 13:32
Last Modified : 11 Dec 2018 11:23
URI: http://epubs.surrey.ac.uk/id/eprint/842176

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800