Esther Lin   林詠惠

Esther Lin profile photo

I am a Computer Science PhD candidate at the University of Toronto, co-advised by Kyros Kutulakos and David Lindell. Previously at Meta Reality Labs, Adobe and Tesla, I worked on modelling and evaluating sensing systems.

I'm interested in pushing optics beyond its fundamental constraints, such as the diffraction limit. I have also built tools for sim-to-real, data generation, design and modelling, as well as agents-in-the-loop that orchestrate them at scale.

My research is generously supported by the SPIE Optical Design and Engineering Scholarship and the NSERC CGS D Scholarship.



Research

NeuralDMD schematic
Ali SaraerToosi, Renbo Tu, Esther Y. H. Lin, Kamyar Azizzadenesheli, Aviad Levis
ECCV 2026
An interpretable neural representation for reconstructing and forecasting spatio-temporal dynamics from sparse and indirect observations.
Superoscillatory point spread functions
Esther Y. H. Lin, Sophia Yang, George V. Eleftheriades, David B. Lindell, Kiriakos N. Kutulakos
ICCP 2026
A framework for designing superoscillatory point spread functions for passive sub-diffraction imaging.
Synthetic 3D eye render for eye tracker prototyping
Esther Y. H. Lin, Yimin Ding, Jogendra Kundu, Yatong An, Mohamed T. El-Haddad, Alexander Fix
ETRA 2026 Short Paper
A framework for evaluating ML-based eye tracking performance for any hardware design using synthetic 3D eye reconstructions.
Lens blur field: point spread functions across the image
Esther Y. H. Lin, Zhecheng Wang, Rebecca Lin, Daniel Miau, Florian Kainz, Jiawen Chen, Cecilia Zhang, David B. Lindell, Kiriakos N. Kutulakos
IEEE TPAMI 2025
A neural differentiable representation for modelling real-world, per-device optical blur.
Super-resolved imaging with adaptive optics pipeline
Robin Swanson, Esther Y. H. Lin, Masen Lamb, Suresh Sivanandam, Kiriakos N. Kutulakos
ICCV 2025 — Highlight
Leverages a telescope's adaptive-optics system to apply learned micro-shifts that, once fused, boost resolution + SNR without altering hardware or routine operation.
Super-oscillatory versus diffraction-limited imaging
Haitang Yang, Esther Y. H. Lin, Kiriakos N. Kutulakos, George V. Eleftheriades
Optica 2022
A theory for incoherent super-oscillatory image formation generalized to any system of lenses for easy integration with existing microscopic or telescopic modalities.
Incoherent super-oscillatory point spread function
Haitang Yang, Esther Y. H. Lin, Kiriakos N. Kutulakos, George V. Eleftheriades
ACS Photonics 2022
Design constraints for super-oscillatory point spread functions can be significantly relaxed in an incoherent system with the convolutional-based image formation model.

Experience


Service


Projects

Laser-driven plume from a thin film Shooting lasers at thin films and droplets
From my internship at the Max Planck Institute for the Structure and Dynamics of Matter, Dr. Dwayne Miller's Atomically Resolved Dynamics group.

Teaching


Fun

Animated Toronto skyline Procrastinating at my desk
An ongoing dump for fun projects :)
Panda drawing from Coding Club Coding Club
A collection of drawings and animations my students and I have made in Coding Club @ Richmond City Centre Community Centre.
Rendered CAD model for optical setups CAD Models for Imaging
A collection of generic components for rendering optical setups. My brother Sam has a more diverse collection.

and...


Updated Nov 15 2026
Thanks, Rebecca for the template!