Hidden
STED — 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 |
|---|---|---|
| depletion_power | -7.0 – 23.0 | % |
| donut_alignment | -7.0 – 23.0 | nm |
| saturation_intensity | -5.6 – 18.4 | % |
Hidden Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | Restormer+ + gradient | 0.751 | 31.97 | 0.942 | 0.77 | ✓ Certified | Zamir et al., ICCV 2024 |
| 2 | DeconvFormer + gradient | 0.721 | 29.02 | 0.901 | 0.85 | ✓ Certified | Chen et al., CVPR 2024 |
| 3 | ScoreMicro + gradient | 0.712 | 29.55 | 0.91 | 0.76 | ✓ Certified | Wei et al., ECCV 2025 |
| 4 | Restormer + gradient | 0.700 | 27.67 | 0.874 | 0.87 | ✓ Certified | Zamir et al., CVPR 2022 |
| 5 | PnP-DnCNN + gradient | 0.663 | 26.5 | 0.846 | 0.8 | ✓ Certified | Zhang et al., IEEE TIP 2017 |
| 6 | ResUNet + gradient | 0.657 | 25.45 | 0.816 | 0.88 | ✓ Certified | DeCelle et al., Nat. Methods 2021 |
| 7 | DiffDeconv + gradient | 0.656 | 26.31 | 0.841 | 0.78 | ✓ Certified | Huang et al., NeurIPS 2024 |
| 8 | TV-Deconvolution + gradient | 0.640 | 25.7 | 0.823 | 0.77 | ✓ Certified | Rudin et al., Phys. A 1992 |
| 9 | PnP-FISTA + gradient | 0.627 | 24.42 | 0.783 | 0.85 | ✓ Certified | Bai et al., 2020 |
| 10 | Wiener Filter + gradient | 0.618 | 24.61 | 0.79 | 0.78 | ✓ Certified | Analytical baseline |
| 11 | U-Net + gradient | 0.599 | 23.91 | 0.765 | 0.77 | ✓ Certified | Ronneberger et al., MICCAI 2015 |
| 12 | Richardson-Lucy + gradient | 0.598 | 23.43 | 0.748 | 0.82 | ✓ Certified | Richardson, JOSA 1972 / Lucy, AJ 1974 |
| 13 | CARE + gradient | 0.579 | 22.49 | 0.711 | 0.85 | ✓ 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%