Hidden

Widefield — Hidden Tier

(5 scenes)

Fully blind server-side evaluation — no data download.

What you get

No data downloadable. Algorithm runs server-side on hidden measurements.

How to use

Package algorithm as Docker container / Python script. Submit via link.

What to submit

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Parameter Specifications

🔒

True spec hidden — blind evaluation, only ranges available.

Parameter Spec Range Unit
psf_sigma -7.0 – 23.0 %
defocus -0.35 – 1.15 μm
background -35.0 – 115.0

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 Restormer + gradient 0.733 29.63 0.911 0.86 ✓ Certified Zamir et al., CVPR 2022
2 DiffDeconv + gradient 0.720 29.83 0.914 0.78 ✓ Certified Huang et al., NeurIPS 2024
3 Restormer+ + gradient 0.702 28.66 0.894 0.79 ✓ Certified Zamir et al., ICCV 2024
4 ResUNet + gradient 0.672 27.37 0.867 0.76 ✓ Certified DeCelle et al., Nat. Methods 2021
5 ScoreMicro + gradient 0.657 25.83 0.827 0.84 ✓ Certified Wei et al., ECCV 2025
6 DeconvFormer + gradient 0.652 26.44 0.844 0.75 ✓ Certified Chen et al., CVPR 2024
7 TV-Deconvolution + gradient 0.640 25.3 0.812 0.81 ✓ Certified Rudin et al., Phys. A 1992
8 Wiener Filter + gradient 0.608 23.69 0.757 0.84 ✓ Certified Analytical baseline
9 U-Net + gradient 0.606 23.35 0.745 0.87 ✓ Certified Ronneberger et al., MICCAI 2015
10 PnP-DnCNN + gradient 0.578 23.07 0.734 0.77 ✓ Certified Zhang et al., IEEE TIP 2017
11 PnP-FISTA + gradient 0.571 22.95 0.729 0.75 ✓ Certified Bai et al., 2020
12 Richardson-Lucy + gradient 0.558 22.29 0.702 0.77 ✓ Certified Richardson, JOSA 1972 / Lucy, AJ 1974
13 CARE + gradient 0.544 20.98 0.645 0.88 ✓ Certified Weigert et al., Nat. Methods 2018

Dataset

Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Widefield