Improving 3D inline computational imaging of textureless objects using pattern illumination

Nicholas Baraghini; Pablo Eugui; Laurin Ginner; Nicole Brosch; Nicholas Baraghini

doi:10.1007/978-3-031-44137-0_34

Improving 3D inline computational imaging of textureless objects using pattern illumination

Nicholas Baraghini, Pablo Eugui, Laurin Ginner, Nicole Brosch, Nicholas Baraghini (Speaker)

High-Performance Vision Systems

University of Bologna

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

Abstract

Feature-based 3D reconstruction methods only work reliably for images with enough features (i.e., texture) that can be matched to infer a depth map. Contradicting the core assumption of such methods, the 3D reconstruction of objects with textureless surfaces remains challenging. This paper explores a simple solution to this problem, i.e., adding artificial texture to such objects. In particular, we equipped a multi-view stereo based inline computational imaging system with a pattern illumination module to compensate for the absence of texture. Comparisons of 3D reconstructions from acquisitions with and without projected patterns show an increase in accuracy when using the pattern illumination.

Original language	English
Title of host publication	Computer Vision Systems
Subtitle of host publication	Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023)
Editors	Henrik I. Christensen, Peter Corke, Renaud Detry, Jean-Baptiste Weibel, Markus Vincze
Pages	412-421
Number of pages	10
Volume	14253
ISBN (Electronic)	978-3-031-44137-0
DOIs	https://doi.org/10.1007/978-3-031-44137-0_34
Publication status	Published - 21 Sept 2023
Event	14th International Conference on Computer Vision Systems (ICVS 2023) - Vienna University of Technology, Vienna, Austria Duration: 27 Sept 2023 → 29 Sept 2023 https://icvs2023.conf.tuwien.ac.at/

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer Cham
Volume	14253
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th International Conference on Computer Vision Systems (ICVS 2023)
Country/Territory	Austria
City	Vienna
Period	27/09/23 → 29/09/23
Internet address	https://icvs2023.conf.tuwien.ac.at/

Research Field

High-Performance Vision Systems

Keywords

3D reconstruction
textureless objects
Inline Computational Imaging
pattern illumination

Access to Document

10.1007/978-3-031-44137-0_34

Cite this

Baraghini, N., Eugui, P., Ginner, L., Brosch, N., & Baraghini, N. (2023). Improving 3D inline computational imaging of textureless objects using pattern illumination. In H. I. Christensen, P. Corke, R. Detry, J.-B. Weibel, & M. Vincze (Eds.), Computer Vision Systems: Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023) (Vol. 14253, pp. 412-421). (Lecture Notes in Computer Science; Vol. 14253). https://doi.org/10.1007/978-3-031-44137-0_34

Baraghini, Nicholas ; Eugui, Pablo ; Ginner, Laurin et al. / Improving 3D inline computational imaging of textureless objects using pattern illumination. Computer Vision Systems: Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023). editor / Henrik I. Christensen ; Peter Corke ; Renaud Detry ; Jean-Baptiste Weibel ; Markus Vincze. Vol. 14253 2023. pp. 412-421 (Lecture Notes in Computer Science).

@inproceedings{381e7f956a9d4d1c975e4139eba03b36,

title = "Improving 3D inline computational imaging of textureless objects using pattern illumination",

abstract = "Feature-based 3D reconstruction methods only work reliably for images with enough features (i.e., texture) that can be matched to infer a depth map. Contradicting the core assumption of such methods, the 3D reconstruction of objects with textureless surfaces remains challenging. This paper explores a simple solution to this problem, i.e., adding artificial texture to such objects. In particular, we equipped a multi-view stereo based inline computational imaging system with a pattern illumination module to compensate for the absence of texture. Comparisons of 3D reconstructions from acquisitions with and without projected patterns show an increase in accuracy when using the pattern illumination.",

keywords = "3D reconstruction, textureless objects, Inline Computational Imaging, pattern illumination",

author = "Nicholas Baraghini and Pablo Eugui and Laurin Ginner and Nicole Brosch and Nicholas Baraghini",

year = "2023",

month = sep,

day = "21",

doi = "10.1007/978-3-031-44137-0_34",

language = "English",

isbn = "978-3-031-44136-3",

volume = "14253",

series = "Lecture Notes in Computer Science",

publisher = "Springer Cham",

pages = "412--421",

editor = "Christensen, {Henrik I.} and Peter Corke and Renaud Detry and Jean-Baptiste Weibel and Markus Vincze",

booktitle = "Computer Vision Systems",

note = "14th International Conference on Computer Vision Systems (ICVS 2023) ; Conference date: 27-09-2023 Through 29-09-2023",

url = "https://icvs2023.conf.tuwien.ac.at/",

}

Baraghini, N , Eugui, P , Ginner, L , Brosch, N & Baraghini, N 2023, Improving 3D inline computational imaging of textureless objects using pattern illumination. in HI Christensen, P Corke, R Detry, J-B Weibel & M Vincze (eds), Computer Vision Systems: Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023). vol. 14253, Lecture Notes in Computer Science, vol. 14253, pp. 412-421, 14th International Conference on Computer Vision Systems (ICVS 2023), Vienna, Austria, 27/09/23. https://doi.org/10.1007/978-3-031-44137-0_34

Improving 3D inline computational imaging of textureless objects using pattern illumination. / Baraghini, Nicholas ; Eugui, Pablo ; Ginner, Laurin et al.
Computer Vision Systems: Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023). ed. / Henrik I. Christensen; Peter Corke; Renaud Detry; Jean-Baptiste Weibel; Markus Vincze. Vol. 14253 2023. p. 412-421 (Lecture Notes in Computer Science; Vol. 14253).

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

TY - GEN

T1 - Improving 3D inline computational imaging of textureless objects using pattern illumination

AU - Baraghini, Nicholas

AU - Eugui, Pablo

AU - Ginner, Laurin

AU - Brosch, Nicole

A2 - Baraghini, Nicholas

A2 - Christensen, Henrik I.

A2 - Corke, Peter

A2 - Detry, Renaud

A2 - Weibel, Jean-Baptiste

A2 - Vincze, Markus

PY - 2023/9/21

Y1 - 2023/9/21

N2 - Feature-based 3D reconstruction methods only work reliably for images with enough features (i.e., texture) that can be matched to infer a depth map. Contradicting the core assumption of such methods, the 3D reconstruction of objects with textureless surfaces remains challenging. This paper explores a simple solution to this problem, i.e., adding artificial texture to such objects. In particular, we equipped a multi-view stereo based inline computational imaging system with a pattern illumination module to compensate for the absence of texture. Comparisons of 3D reconstructions from acquisitions with and without projected patterns show an increase in accuracy when using the pattern illumination.

AB - Feature-based 3D reconstruction methods only work reliably for images with enough features (i.e., texture) that can be matched to infer a depth map. Contradicting the core assumption of such methods, the 3D reconstruction of objects with textureless surfaces remains challenging. This paper explores a simple solution to this problem, i.e., adding artificial texture to such objects. In particular, we equipped a multi-view stereo based inline computational imaging system with a pattern illumination module to compensate for the absence of texture. Comparisons of 3D reconstructions from acquisitions with and without projected patterns show an increase in accuracy when using the pattern illumination.

KW - 3D reconstruction

KW - textureless objects

KW - Inline Computational Imaging

KW - pattern illumination

U2 - 10.1007/978-3-031-44137-0_34

DO - 10.1007/978-3-031-44137-0_34

M3 - Conference Proceedings with Oral Presentation

SN - 978-3-031-44136-3

VL - 14253

T3 - Lecture Notes in Computer Science

SP - 412

EP - 421

BT - Computer Vision Systems

T2 - 14th International Conference on Computer Vision Systems (ICVS 2023)

Y2 - 27 September 2023 through 29 September 2023

ER -

Baraghini N , Eugui P , Ginner L , Brosch N , Baraghini N. Improving 3D inline computational imaging of textureless objects using pattern illumination. In Christensen HI, Corke P, Detry R, Weibel JB, Vincze M, editors, Computer Vision Systems: Proceedings of the 14th International Conference on Computer Vision Systems (ICVS 2023). Vol. 14253. 2023. p. 412-421. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-44137-0_34

Improving 3D inline computational imaging of textureless objects using pattern illumination

Abstract

Publication series

Conference

Research Field

Keywords

Access to Document

Fingerprint

Cite this