Joint Optical Design for Computer Vision Tasks: Challenges and Solutions

Geoffroi Côté; Simon Thibault; Jean François Lalonde; Felix Heide

doi:10.1117/12.2691993

Joint Optical Design for Computer Vision Tasks: Challenges and Solutions

Geoffroi Côté, Simon Thibault, Jean François Lalonde, Felix Heide

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

This paper delves into the substantial challenges faced in end-to-end computer vision applications where optical systems are jointly optimized alongside downstream data-driven models, and explores available solutions.

Original language	English (US)
Title of host publication	International Optical Design Conference 2023
Editors	David M. Aikens, Henning Rehn, Simon Thibault, Kristina Uhlendorf
Publisher	SPIE
ISBN (Electronic)	9781510668508
DOIs	https://doi.org/10.1117/12.2691993
State	Published - 2023
Event	International Optical Design Conference 2023 - Quebec City, Canada Duration: Jun 4 2023 → Jun 8 2023

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	12798
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	International Optical Design Conference 2023
Country/Territory	Canada
City	Quebec City
Period	6/4/23 → 6/8/23

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Keywords

computer vision
deep learning
end-to-end optimization
optical design

Access to Document

10.1117/12.2691993

Cite this

Côté, G., Thibault, S., Lalonde, J. F., & Heide, F. (2023). Joint Optical Design for Computer Vision Tasks: Challenges and Solutions. In D. M. Aikens, H. Rehn, S. Thibault, & K. Uhlendorf (Eds.), International Optical Design Conference 2023 Article 127980S (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 12798). SPIE. https://doi.org/10.1117/12.2691993

@inproceedings{e959720644e44002b572b3f22bed719a,

title = "Joint Optical Design for Computer Vision Tasks: Challenges and Solutions",

abstract = "This paper delves into the substantial challenges faced in end-to-end computer vision applications where optical systems are jointly optimized alongside downstream data-driven models, and explores available solutions.",

keywords = "computer vision, deep learning, end-to-end optimization, optical design",

author = "Geoffroi C{\^o}t{\'e} and Simon Thibault and Lalonde, {Jean Fran{\c c}ois} and Felix Heide",

year = "2023",

doi = "10.1117/12.2691993",

language = "English (US)",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Aikens, {David M.} and Henning Rehn and Simon Thibault and Kristina Uhlendorf",

booktitle = "International Optical Design Conference 2023",

address = "United States",

}

Côté, G, Thibault, S, Lalonde, JF & Heide, F 2023, Joint Optical Design for Computer Vision Tasks: Challenges and Solutions. in DM Aikens, H Rehn, S Thibault & K Uhlendorf (eds), International Optical Design Conference 2023., 127980S, Proceedings of SPIE - The International Society for Optical Engineering, vol. 12798, SPIE, International Optical Design Conference 2023, Quebec City, Canada, 6/4/23. https://doi.org/10.1117/12.2691993

Joint Optical Design for Computer Vision Tasks: Challenges and Solutions. / Côté, Geoffroi; Thibault, Simon; Lalonde, Jean François et al.
International Optical Design Conference 2023. ed. / David M. Aikens; Henning Rehn; Simon Thibault; Kristina Uhlendorf. SPIE, 2023. 127980S (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 12798).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Joint Optical Design for Computer Vision Tasks

T2 - International Optical Design Conference 2023

AU - Côté, Geoffroi

AU - Thibault, Simon

AU - Lalonde, Jean François

AU - Heide, Felix

PY - 2023

Y1 - 2023

N2 - This paper delves into the substantial challenges faced in end-to-end computer vision applications where optical systems are jointly optimized alongside downstream data-driven models, and explores available solutions.

AB - This paper delves into the substantial challenges faced in end-to-end computer vision applications where optical systems are jointly optimized alongside downstream data-driven models, and explores available solutions.

KW - computer vision

KW - deep learning

KW - end-to-end optimization

KW - optical design

UR - http://www.scopus.com/inward/record.url?scp=85174164954&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85174164954&partnerID=8YFLogxK

U2 - 10.1117/12.2691993

DO - 10.1117/12.2691993

M3 - Conference contribution

AN - SCOPUS:85174164954

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - International Optical Design Conference 2023

A2 - Aikens, David M.

A2 - Rehn, Henning

A2 - Thibault, Simon

A2 - Uhlendorf, Kristina

PB - SPIE

Y2 - 4 June 2023 through 8 June 2023

ER -

Joint Optical Design for Computer Vision Tasks: Challenges and Solutions

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this