Joshua Chen

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2024-110

May 16, 2024

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-110.pdf

Millions of people across the world have visual aberrations that prevent them from using their digital devices without corrective eyewear. Vision correcting displays aim to present an in-focus image to the user without the use of such eyewear. This work proposes two new methods for performing computational vision correction. The first method builds upon existing research utilizing compressive sampling for image deconvolution. The second method utilizes a Vision Transformer-based model to perform image deconvolution. These two methods are presented and evaluated against previous methods. Lastly, future research directions are suggested that could improve upon the methods in this work and bring vision-correcting displays closer to a practical application that millions of people can use.

Advisors: Brian A. Barsky

---

BibTeX citation:

@mastersthesis{Chen:EECS-2024-110,
    Author= {Chen, Joshua},
    Title= {Towards Fast and Accurate Computational Algorithms for Vision Correcting Displays},
    School= {EECS Department, University of California, Berkeley},
    Year= {2024},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-110.html},
    Number= {UCB/EECS-2024-110},
    Abstract= {Millions of people across the world have visual aberrations that prevent them from using their digital devices without corrective eyewear. Vision correcting displays aim to present an in-focus image to the user without the use of such eyewear. This work proposes two new methods for performing computational vision correction. The first method builds upon existing research utilizing compressive sampling for image deconvolution. The second method utilizes a Vision Transformer-based model to perform image deconvolution. These two methods are presented and evaluated against previous methods. Lastly, future research directions are suggested that could improve upon the methods in this work and bring vision-correcting displays closer to a practical application that millions of people can use.},
}

EndNote citation:

%0 Thesis
%A Chen, Joshua 
%T Towards Fast and Accurate Computational Algorithms for Vision Correcting Displays
%I EECS Department, University of California, Berkeley
%D 2024
%8 May 16
%@ UCB/EECS-2024-110
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-110.html
%F Chen:EECS-2024-110