A Fourier Neural Operator trained at 128x128 reconstructs 256x256 and 512x512 lensless diffuser images with under 1 dB PSNR loss and outperforms a comparable U-Net by 2.14 dB PSNR.
Super-resolution neural operator
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.optics 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Resolution-Agnostic Lensless Imaging via Fourier Neural Operators
A Fourier Neural Operator trained at 128x128 reconstructs 256x256 and 512x512 lensless diffuser images with under 1 dB PSNR loss and outperforms a comparable U-Net by 2.14 dB PSNR.