2010
DOI: 10.1111/j.1467-8659.2009.01617.x
|View full text |Cite
|
Sign up to set email alerts
|

Seamless Montage for Texturing Models

Abstract: We present an automatic method to recover high-resolution texture over an object by mapping detailed photographs onto its surface. Such high-resolution detail often reveals inaccuracies in geometry and registration, as well as lighting variations and surface reflections. Simple image projection results in visible seams on the surface. We minimize such seams using a global optimization that assigns compatible texture to adjacent triangles. The key idea is to search not only combinatorially over the source image… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
68
0
5

Year Published

2012
2012
2023
2023

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 115 publications
(73 citation statements)
references
References 22 publications
0
68
0
5
Order By: Relevance
“…This is essentially due to the low resolution of the cameras, and to the difficulty of handling the peculiar illumination provided by the projector. Other systems have been proposed which take into account also the color, but they are not able to achieve real-time performances [6] or to reconstruct the geometry in an accurate way [7].…”
Section: A Real-time 3d Scanningmentioning
confidence: 99%
“…This is essentially due to the low resolution of the cameras, and to the difficulty of handling the peculiar illumination provided by the projector. Other systems have been proposed which take into account also the color, but they are not able to achieve real-time performances [6] or to reconstruct the geometry in an accurate way [7].…”
Section: A Real-time 3d Scanningmentioning
confidence: 99%
“…The cost function is designed so that light probes with a good resolution in the direction of the pixel are preferred and uses smoothness constraints to avoid visible seams in the resulting textures. The approach is similar to previous methods for high quality texture projection [68,131], but extended to handle omnidirectional source images with considerable distortion in angles close to the backward facing direction.…”
Section: Reconstructing the Scene Modelmentioning
confidence: 94%
“…That means the detail-rich information which is contained in high-resolution photographic images, is completely transmitted to the reconstructed models. Our technique is based on the techniques presented in (Allene et al, 2008), (Gal et al, 2010) and (Lempitsky and Ivanov, 2007) with a few modifications for the energy function: The first term, which measures the visual details of each face, is computed as a combination of a number of visible pixels of a given face and the angle between the face's normal and the looking vector of the camera. The second term, which measures the color continuity between adjacent faces, is computed as difference of the average RGB color per triangle.…”
Section: Discussionmentioning
confidence: 99%
“…The authors in (Allene et al, 2008) proposed to calculate the data term related to the projection area of the 3D triangle in image space. A method in (Gal et al, 2010) introduces the assignment of the faces with a set of transformed images, which compensate the geometric errors, and, applying Poisson blending, removes lighting variations. The methodology proposed in this paper is based on the Markov Random Field approach which allows not only the selection of one best view per face using a special criterion, but also simultaneously apply regularization on the texture assignment to minimize seam visibilities on the face boundaries.…”
Section: Related Workmentioning
confidence: 99%