Reproduce “Image Reconstruction from Phase with Untrained Neural Priors” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ene Meco, Ahmet Enis Cetin
The authors present a phase-only microscopy reconstruction method that alternates Fourier-phase and image-support constraints with an image-specific neural prior, then refines using constraints alone. In a noise-free, known-support evaluation on 77 paired images, SelfDeblur achieved 31.41 dB pooled PSNR and 7.24 × 10⁻⁴ pooled MSE after 500 refinement passes, outperforming the listed constraint-only row; phase does not determine absolute intensity.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.