Deep learning-enabled framework for automatic lens design starting point generation
Authors: | Côté, Geoffroi; Lalonde, Jean-François; Thibault, Simon |
Abstract: | We present a simple, highly modular deep neural network (DNN) framework to address the problem of automatically inferring lens design starting points tailored to the desired specifications. In contrast to previous work, our model can handle various and complex lens structures suitable for real-world problems such as Cooke Triplets or Double Gauss lenses. Our successfully trained dynamic model can infer lens designs with realistic glass materials whose optical performance compares favorably to reference designs from the literature on 80 different lens structures. Using our trained model as a backbone, we make available to the community a web application that outputs a selection of varied, high-quality starting points directly from the desired specifications, which we believe will complement any lens designer’s toolbox. |
Document Type: | Article de recherche |
Issue Date: | 25 January 2021 |
Open Access Date: | 18 February 2021 |
Document version: | VoR |
Creative Commons Licence: | https://creativecommons.org/licenses/by/4.0 |
Permalink: | http://hdl.handle.net/20.500.11794/68199 |
This document was published in: | Optics express, Vol. 29 (3), 3841-3854 (2021) https://doi.org/10.1364/OE.401590 Optical Society of America |
Alternative version: | 10.1364/OE.401590 |
Collection: | Articles publiés dans des revues avec comité de lecture |
Files in this item:
Description | Size | Format | ||
---|---|---|---|---|
oe-29-3-3841.pdf | 2.1 MB | Adobe PDF | ![]() View/Open |
All documents in CorpusUL are protected by Copyright Act of Canada.