Automatic Lens Design based on Differentiable Ray-tracing

Qiang Fu; Wolfgang Heidrich

doi:10.1364/cosi.2022.cth4c.2

Automatic Lens Design based on Differentiable Ray-tracing

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

Abstract

We propose a fully differentiable optical design method enabled by curriculum learning. Preliminary results show that our framework is suitable to solve highly non-convex problems like cellphone lens design.

Original language	English (US)
Title of host publication	Imaging and Applied Optics Congress 2022 (3D, AOA, COSI, ISA, pcAOP)
Publisher	Optica Publishing Group
ISBN (Print)	9781557528209
DOIs	https://doi.org/10.1364/cosi.2022.cth4c.2
State	Published - 2022

Access to Document

10.1364/cosi.2022.cth4c.2

https://repository.kaust.edu.sa/bitstream/10754/685232/1/OSAmeetings.pdf

Cite this

@inproceedings{2687f14f52b842dd901fb0b23f4cab06,

title = "Automatic Lens Design based on Differentiable Ray-tracing",

abstract = "We propose a fully differentiable optical design method enabled by curriculum learning. Preliminary results show that our framework is suitable to solve highly non-convex problems like cellphone lens design.",

author = "Qiang Fu and Wolfgang Heidrich",

note = "KAUST Repository Item: Exported on 2022-10-31",

year = "2022",

doi = "10.1364/cosi.2022.cth4c.2",

language = "English (US)",

isbn = "9781557528209",

booktitle = "Imaging and Applied Optics Congress 2022 (3D, AOA, COSI, ISA, pcAOP)",

publisher = "Optica Publishing Group",

}

TY - GEN

T1 - Automatic Lens Design based on Differentiable Ray-tracing

AU - Fu, Qiang

AU - Heidrich, Wolfgang

N1 - KAUST Repository Item: Exported on 2022-10-31

PY - 2022

Y1 - 2022

N2 - We propose a fully differentiable optical design method enabled by curriculum learning. Preliminary results show that our framework is suitable to solve highly non-convex problems like cellphone lens design.

AB - We propose a fully differentiable optical design method enabled by curriculum learning. Preliminary results show that our framework is suitable to solve highly non-convex problems like cellphone lens design.

UR - http://hdl.handle.net/10754/685232

UR - https://opg.optica.org/abstract.cfm?URI=COSI-2022-CTh4C.2

UR - http://www.scopus.com/inward/record.url?scp=85139118108&partnerID=8YFLogxK

U2 - 10.1364/cosi.2022.cth4c.2

DO - 10.1364/cosi.2022.cth4c.2

M3 - Conference contribution

SN - 9781557528209

BT - Imaging and Applied Optics Congress 2022 (3D, AOA, COSI, ISA, pcAOP)

PB - Optica Publishing Group

ER -

Automatic Lens Design based on Differentiable Ray-tracing

Abstract

Access to Document

Other files and links

Fingerprint

Cite this