inria-00410980, version 1

Voir la fiche détaillée

BibTeX  EndNote  TEI  RefWorks %% inria-00410980, version 1 %% http://hal.inria.fr/inria-00410980/en/ @article{SALMAN:2009:INRIA-00410980:1, title = { {H}igh {R}esolution {S}urface {R}econstruction from {O}verlapping {M}ultiple-{V}iews}, author = {{S}alman, {N}ader and {Y}vinec, {M}ariette}, abstract = {{E}xtracting a computer model of a real scene from a sequence of views, is one of the most challenging and fundamental problems in computer vision. {S}tereo vision algorithms allow us to extract from the images a sparse 3{D} point cloud on the scene surfaces. {H}owever, computing an accurate mesh of the scene based on such poor quality data points (noise, sparsity) is very difficult. {H}ere we describe a simple yet original approach that uses both the stereo vision extracted point cloud and the calibrated images. {O}ur method is a three-stage process in which the first stage merges, filters and smoothes the input 3{D} points. {T}he second stage builds for each calibrated image a triangular depth-map and fuses the set of depth-maps into a triangle soup that minimize violations of size and visibility constraints. {F}inally, a mesh is computed from the triangle soup using a reconstruction method that combines restricted {D}elaunay triangulation and {D}elaunay refinement.}, keywords = {{M}ultiple-views; {S}urface reconstruction; {T}riangle soup; {R}estricted {D}elaunay triangulation; {D}elaunay refinement; {C}gal; {CI}mg}, language = {{A}nglais}, affiliation = {{GEOMETRICA} - {INRIA} {S}ophia {A}ntipolis / {INRIA} {S}aclay - {I}le de {F}rance - {INRIA} }, booktitle = {{P}roceedings of the 25th annual symposium on {C}omputational geometry }, publisher = {{ACM} {N}ew {Y}ork, {NY}, {USA} }, pages = {104-105 }, journal = {{A}nnual {S}ymposium on {C}omputational {G}eometry }, note = {{I}.: {C}omputing {M}ethodologies/{I}.3: {COMPUTER} {GRAPHICS}/{I}.3.5: {C}omputational {G}eometry and {O}bject {M}odeling, {I}.: {C}omputing {M}ethodologies/{I}.3: {COMPUTER} {GRAPHICS}/{I}.3.8: {A}pplications, {I}.: {C}omputing {M}ethodologies/{I}.4: {IMAGE} {PROCESSING} {AND} {COMPUTER} {VISION}/{I}.4.5: {R}econstruction }, audience = {internationale }, doi = {10.1145/1542362.1542386 }, day = {08}, month = {06}, year = {2009}, URL = {http://dx.doi.org/10.1145/1542362.1542386 http://hal.inria.fr/inria-00410980/en/}, URL = {http://hal.inria.fr/inria-00410980/PDF/p104-salman.pdf}, } %F inria-00410980, version 1 %0 Journal Article %U http://hal.inria.fr/inria-00410980/en/ %2 INFO:INFO_CG %T High Resolution Surface Reconstruction from Overlapping Multiple-Views %A Salman, Nader %A Yvinec, Mariette %A Salman, N. %A Yvinec, M. %A Salman N. et al %+ GEOMETRICA - INRIA Sophia Antipolis / INRIA Saclay - Ile de France - INRIA %X Extracting a computer model of a real scene from a sequence of views, is one of the most challenging and fundamental problems in computer vision. Stereo vision algorithms allow us to extract from the images a sparse 3D point cloud on the scene surfaces. However, computing an accurate mesh of the scene based on such poor quality data points (noise, sparsity) is very difficult. Here we describe a simple yet original approach that uses both the stereo vision extracted point cloud and the calibrated images. Our method is a three-stage process in which the first stage merges, filters and smoothes the input 3D points. The second stage builds for each calibrated image a triangular depth-map and fuses the set of depth-maps into a triangle soup that minimize violations of size and visibility constraints. Finally, a mesh is computed from the triangle soup using a reconstruction method that combines restricted Delaunay triangulation and Delaunay refinement. %2 I.: Computing Methodologies/I.3: COMPUTER GRAPHICS/I.3.5: Computational Geometry and Object Modeling, I.: Computing Methodologies/I.3: COMPUTER GRAPHICS/I.3.8: Applications, I.: Computing Methodologies/I.4: IMAGE PROCESSING AND COMPUTER VISION/I.4.5: Reconstruction %G Anglais %J Annual Symposium on Computational Geometry %8 2009-06-08 %2 internationale %I ACM New York, NY, USA %2 Proceedings of the 25th annual symposium on Computational geometry %P 104-105 %R 10.1145/1542362.1542386 %K Multiple-views; Surface reconstruction; Triangle soup; Restricted Delaunay triangulation; Delaunay refinement; Cgal; CImg %2 Gyroviz, ANR-07- AM-013 <?xml version="1.0" encoding="UTF-8"?> <listBibl xmlns="http://www.tei-c.org/ns/1.0" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <listBibl type="article"><head>Articles dans des revues avec comité de lecture</head> <biblStruct type="article" xml:id="inria-00410980"><analytic> <author><persName><forename>Nader</forename><surname>Salman</surname></persName> <affiliation> <orgName type="laboratory">INRIA Sophia Antipolis / INRIA Saclay - Ile de France</orgName> <orgName type="team">GEOMETRICA</orgName> <country>FR</country> <orgName type="institution">INRIA</orgName> </affiliation> </author> <author><persName><forename>Mariette</forename><surname>Yvinec</surname></persName> <affiliation> <orgName type="laboratory">INRIA Sophia Antipolis / INRIA Saclay - Ile de France</orgName> <orgName type="team">GEOMETRICA</orgName> <country>FR</country> <orgName type="institution">INRIA</orgName> </affiliation> </author> <title type="main" level="a">High Resolution Surface Reconstruction from Overlapping Multiple-Views</title></analytic><monogr><title level="j" type="main">Annual Symposium on Computational Geometry</title><imprint><publisher>ACM New York, NY, USA</publisher><biblScope type="publicationDate">2009-06-08</biblScope><biblScope type="pp">104-105</biblScope></imprint></monogr><idno type="doi">10.1145/1542362.1542386</idno><idno type="HALid">http://hal.inria.fr/inria-00410980/en/</idno><idno type="HALFile">http://hal.inria.fr/inria-00410980/PDF/p104-salman.pdf</idno></biblStruct> </listBibl> </listBibl>

High Resolution Surface Reconstruction from Overlapping Multiple-Views Nader Salman (  Auteur à contacter de préférence , ) 1, Mariette Yvinec (, ) a, 1 (08/06/2009)

Accès au texte intégral PDF

Annual Symposium on Computational Geometry (2009) 104-105

Lien vers texte intégral chez l'éditeur

Résumé : Extracting a computer model of a real scene from a sequence of views, is one of the most challenging and fundamental problems in computer vision. Stereo vision algorithms allow us to extract from the images a sparse 3D point cloud on the scene surfaces. However, computing an accurate mesh of the scene based on such poor quality data points (noise, sparsity) is very difficult. Here we describe a simple yet original approach that uses both the stereo vision extracted point cloud and the calibrated images. Our method is a three-stage process in which the first stage merges, filters and smoothes the input 3D points. The second stage builds for each calibrated image a triangular depth-map and fuses the set of depth-maps into a triangle soup that minimize violations of size and visibility constraints. Finally, a mesh is computed from the triangle soup using a reconstruction method that combines restricted Delaunay triangulation and Delaunay refinement.

a –	INRIA
1 :	GEOMETRICA (INRIA Sophia Antipolis / INRIA Saclay - Ile de France)
	INRIA

Domaine

:

Informatique/Géométrie algorithmique

Mots-clés : Multiple-views – Surface reconstruction – Triangle soup – Restricted Delaunay triangulation – Delaunay refinement – Cgal – CImg

DOI : 10.1145/1542362.1542386

inria-00410980, version 1

http://hal.inria.fr/inria-00410980/fr/

oai:hal.inria.fr:inria-00410980_v1

Contributeur : Nader Salman <>

Soumis le : Mardi 25 Août 2009, 14:31:54

Dernière modification le : Lundi 21 Septembre 2009, 16:00:12