Dev

STED — 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
depletion_power -12.0 – 18.0 %
donut_alignment -12.0 – 18.0 nm
saturation_intensity -9.6 – 14.4 %

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 Restormer+ + gradient 0.789 33.15 0.954 0.88 ✓ Certified Zamir et al., ICCV 2024
2 DeconvFormer + gradient 0.770 32.52 0.948 0.83 ✓ Certified Chen et al., CVPR 2024
3 ScoreMicro + gradient 0.741 30.16 0.919 0.86 ✓ Certified Wei et al., ECCV 2025
4 Restormer + gradient 0.738 30.58 0.925 0.81 ✓ Certified Zamir et al., CVPR 2022
5 DiffDeconv + gradient 0.721 28.58 0.892 0.89 ✓ Certified Huang et al., NeurIPS 2024
6 PnP-DnCNN + gradient 0.701 28.05 0.882 0.84 ✓ Certified Zhang et al., IEEE TIP 2017
7 ResUNet + gradient 0.696 28.3 0.887 0.79 ✓ Certified DeCelle et al., Nat. Methods 2021
8 PnP-FISTA + gradient 0.690 27.47 0.869 0.84 ✓ Certified Bai et al., 2020
9 TV-Deconvolution + gradient 0.656 25.58 0.82 0.86 ✓ Certified Rudin et al., Phys. A 1992
10 U-Net + gradient 0.649 25.26 0.81 0.86 ✓ Certified Ronneberger et al., MICCAI 2015
11 Wiener Filter + gradient 0.647 25.54 0.819 0.82 ✓ Certified Analytical baseline
12 CARE + gradient 0.643 24.82 0.796 0.88 ✓ Certified Weigert et al., Nat. Methods 2018
13 Richardson-Lucy + gradient 0.621 24.3 0.779 0.83 ✓ 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 STED