Dev

Light-Sheet — Dev Tier

(5 scenes)

Blind evaluation tier — no ground truth available.

What you get

Measurements (y), ideal forward operator (H), and spec ranges only.

How to use

Apply your pipeline from the Public tier. Use consistency as self-check.

What to submit

Reconstructed signals and corrected spec. Scored server-side.

Parameter Specifications

🔒

True spec hidden — estimate parameters from spec ranges below.

Parameter Spec Range Unit
sheet_thickness -1.2 – 1.8 μm
sheet_tilt -0.6 – 0.9 deg
stripe_artifact -0.12 – 0.18

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DeconvFormer + gradient 0.772 32.8 0.951 0.82 ✓ Certified Chen et al., CVPR 2024
2 Restormer+ + gradient 0.771 33.17 0.954 0.79 ✓ Certified Zamir et al., ICCV 2024
3 Restormer + gradient 0.769 31.55 0.938 0.89 ✓ Certified Zamir et al., CVPR 2022
4 ScoreMicro + gradient 0.758 31.1 0.932 0.87 ✓ Certified Wei et al., ECCV 2025
5 DiffDeconv + gradient 0.734 30.32 0.922 0.81 ✓ Certified Huang et al., NeurIPS 2024
6 ResUNet + gradient 0.722 29.19 0.904 0.84 ✓ Certified DeCelle et al., Nat. Methods 2021
7 CARE + gradient 0.700 27.55 0.871 0.88 ✓ Certified Weigert et al., Nat. Methods 2018
8 U-Net + gradient 0.697 28.25 0.886 0.8 ✓ Certified Ronneberger et al., MICCAI 2015
9 PnP-DnCNN + gradient 0.682 26.65 0.849 0.88 ✓ Certified Zhang et al., IEEE TIP 2017
10 TV-Deconvolution + gradient 0.663 26.67 0.85 0.78 ✓ Certified Rudin et al., Phys. A 1992
11 PnP-FISTA + gradient 0.629 24.21 0.776 0.88 ✓ Certified Bai et al., 2020
12 Wiener Filter + gradient 0.606 23.44 0.748 0.86 ✓ Certified Analytical baseline
13 Richardson-Lucy + gradient 0.599 23.17 0.738 0.86 ✓ Certified Richardson, JOSA 1972 / Lucy, AJ 1974

Visible Data Fields

y H_ideal spec_ranges

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 Light-Sheet