Select a date or location to show sessions only on that day or location.
Select a single session for a detailed view (with abstracts and downloads when you are logged in as a registered attendee).
Please note that all times are shown in the time zone of the conference. The current conference time is: 10th June 2025, 08:31:27am CEST
Session Chair: Stefan Bäumer, TNO, Netherlands, The
Location:Collegezaal D
Presentations
8:30am - 9:00am INVITED
AI – Assisted Optical Design: A New Era
Simon Thibault
University Laval, Canada
In this presentation, I will review the latest developments on the use of AI in optical engineering. We'll see how AI can be used for teaching optical engineering, in particular using AI generative tools. Over the last few years, several uses of AI in optics have been published but few developments have really changed the life of an optical engineer. We'll also take a closer look at how generative AI could be used by optical designers. Finally, we'll look to the future and how we want to build the future of optical engineering with these new tools.
9:00am - 9:15am
Designing off-axis augmented reality display systems based on freeform holographic optical elements
Tong Yang, Yongdong Wang, Xin Lyu, Dewen Cheng, Yongtian Wang
Beijing Institute of Technology, China
The design of off-axis augmented reality display systems based on freeform holographic optical elements (HOE) is presented. The complex freeform HOE is fabricated using freeform optics. Joint optimization method of imaging and recording systems is proposed considering the imaging performance, diffraction efficiency and design constraints. Prototypes including near-eye display and head-up display systems are developed. Larger FOV and more compact structure are realized, compared with the systems using traditional HOEs.
9:15am - 9:30am
Imaging optical design based on inverse methods
Sanjana Verma1, Koondanibha Mitra1, Lisa Kusch1, Martijn J.H. Anthonissen1, Jan H.M. ten Thije Boonkkamp1, Wilbert L. IJzerman2,1
1Eindhoven University of Technology, PO Box 513, 5600 MB Eindhoven, The Netherlands; 2Signify Research, High Tech Campus 7, 5656 AE Eindhoven, The Netherlands
We propose a method to design freeform imaging systems using inverse methods from nonimaging optics. A linear optical map in phase space implies that the ratio of source and target energy distributions is a constant. The performance of the inverse freeform design is compared to a classical design by raytracing parallel beams of light and comparing the corresponding spot sizes. The inverse design significantly outperforms the classical design.
9:30am - 9:45am
Constrained optimization of a zoom lens with CMA-ES algorithm
Tristan Marty1,2, Sébastien Héron1, Yann Semet1
1Thales Research and Technology, France; 2Inria and École Polytechnique
In the present paper we investigate how optimization algorithm can be tailored to improve the lens design process. We replaced gradient-based optimisation methods by the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). This stochastic algorithm is considered more robust and is well suited to avoid local optima often found in optical design. In addition, the algorithm is paired with an augmented Lagrangian method to incorporate constraints handling inside the computation framework. Performances are illustrated on a photographic zoom lens.
9:45am - 10:00am
All-Optical Convolution Enabled by Photochromic Media
Alessandro Bile1, Mario Bragaglia2, Francesca Nanni2, Eugenio Fazio1
1Sapienza University of Rome, Italy; 2Department of Enterprise Engineering “Mario Lucertini”, University of Rome Tor Vergata
We present a fully optical system for the convolution of spatial features in photochromic media. The method relies on imprinting the Fourier transform of a mask into a photochromic sample using 405 nm light, and subsequently performing a convolution using a second, non-activating wavelength (470 nm). Our approach offers reconfigurability and dynamic training capabilities, overcoming the rigidity of traditional optical correlators. Experimental and simulated results demonstrate the system’s ability to distinguish spatial features with high selectivity, enabling compact and efficient optical pattern recognition. This work paves the way for new applications in optical signal processing, machine vision, and embedded neuromorphic photonic hardware.
