Dev
Confocal 3D — 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 |
|---|---|---|
| z_step | -60.0 – 90.0 | nm |
| spherical_aberr | -0.12 – 0.18 | waves |
| refractive_index | 1.503 – 1.533 |
Dev Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | SwinIR-3D + gradient | 0.787 | 33.45 | 0.956 | 0.85 | ✓ Certified | Liang et al., ICCV 2021 (3D adapted) |
| 2 | DiffusionMicro + gradient | 0.765 | 31.85 | 0.941 | 0.85 | ✓ Certified | Gao et al., Nat. Methods 2024 |
| 3 | Restormer-3D + gradient | 0.759 | 31.6 | 0.938 | 0.84 | ✓ Certified | Zamir et al., CVPR 2022 (3D adapted) |
| 4 | IRCNN-Confocal + gradient | 0.729 | 29.53 | 0.909 | 0.85 | ✓ Certified | Zhang et al., CVPR 2017 |
| 5 | CARE + gradient | 0.679 | 26.78 | 0.853 | 0.85 | ✓ Certified | Weigert et al., Nat. Methods 2018 |
| 6 | Noise2Void + gradient | 0.667 | 25.99 | 0.832 | 0.87 | ✓ Certified | Krull et al., CVPR 2019 |
| 7 | Wiener-3D + gradient | 0.660 | 26.03 | 0.833 | 0.83 | ✓ Certified | Wiener, 1942 |
| 8 | Richardson-Lucy + gradient | 0.647 | 25.34 | 0.813 | 0.84 | ✓ Certified | Richardson, J. Opt. Soc. Am. 1972 |
| 9 | U-Net-3D + gradient | 0.640 | 24.69 | 0.792 | 0.88 | ✓ Certified | Çiçek et al., MICCAI 2016 |
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%