Public

Widefield — Public Tier

(5 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

✓

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
psf_sigma -10.0 – 20.0 5.0 %
defocus -0.5 – 1.0 0.25 μm
background -50.0 – 100.0 25.0

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

28.76 dB

SSIM 0.8191

Scenario II (Mismatch)

23.41 dB

SSIM 0.5268

Scenario III (Oracle)

7.23 dB

SSIM 0.0107

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 24.90 0.8560 21.08 0.7343 8.80 0.0149
scene_01 30.17 0.7563 23.36 0.3394 8.75 0.0099
scene_02 32.75 0.8090 25.73 0.3127 5.44 0.0063
scene_03 27.21 0.8553 23.48 0.7210 5.93 0.0117

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 Restormer+ + gradient 0.838 36.03 0.974 0.94 ✓ Certified Zamir et al., ICCV 2024
2 DeconvFormer + gradient 0.835 35.81 0.972 0.94 ✓ Certified Chen et al., CVPR 2024
3 ScoreMicro + gradient 0.828 35.83 0.973 0.9 ✓ Certified Wei et al., ECCV 2025
4 DiffDeconv + gradient 0.822 35.39 0.97 0.9 ✓ Certified Huang et al., NeurIPS 2024
5 Restormer + gradient 0.816 34.47 0.964 0.93 ✓ Certified Zamir et al., CVPR 2022
6 ResUNet + gradient 0.816 34.28 0.963 0.94 ✓ Certified DeCelle et al., Nat. Methods 2021
7 U-Net + gradient 0.784 32.62 0.949 0.89 ✓ Certified Ronneberger et al., MICCAI 2015
8 CARE + gradient 0.773 31.58 0.938 0.91 ✓ Certified Weigert et al., Nat. Methods 2018
9 PnP-DnCNN + gradient 0.752 29.94 0.916 0.93 ✓ Certified Zhang et al., IEEE TIP 2017
10 PnP-FISTA + gradient 0.736 28.76 0.896 0.95 ✓ Certified Bai et al., 2020
11 TV-Deconvolution + gradient 0.694 27.13 0.861 0.89 ✓ Certified Rudin et al., Phys. A 1992
12 Wiener Filter + gradient 0.670 26.04 0.833 0.88 ✓ Certified Analytical baseline
13 Richardson-Lucy + gradient 0.644 24.77 0.795 0.89 ✓ Certified Richardson, JOSA 1972 / Lucy, AJ 1974

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
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