Dev
Confocal Live-Cell — 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 |
|---|---|---|
| pinhole | -6.0 – 9.0 | μm |
| refractive_index | 1.509 – 1.524 | |
| photobleaching | -6.0 – 9.0 | % |
Dev Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | Restormer-Micro + gradient | 0.755 | 31.19 | 0.933 | 0.85 | ✓ Certified | Zamir et al., CVPR 2022 (microscopy) |
| 2 | DiffusionCell + gradient | 0.745 | 31.2 | 0.933 | 0.8 | ✓ Certified | Gao et al., Nat. Methods 2024 |
| 3 | SwinIR-LiveCell + gradient | 0.733 | 30.13 | 0.919 | 0.82 | ✓ Certified | Liang et al., ICCV 2021 (live-cell) |
| 4 | PN2V + gradient | 0.616 | 23.69 | 0.757 | 0.88 | ✓ Certified | Krull et al., ECCV 2020 |
| 5 | Noise2Void + gradient | 0.605 | 24.05 | 0.77 | 0.78 | ✓ Certified | Krull et al., CVPR 2019 |
| 6 | CARE + gradient | 0.592 | 22.89 | 0.727 | 0.86 | ✓ Certified | Weigert et al., Nat. Methods 2018 |
| 7 | NLM-Fluorescence + gradient | 0.564 | 21.85 | 0.684 | 0.86 | ✓ Certified | Buades et al., CVPR 2005 |
| 8 | Noise2Self + gradient | 0.527 | 21.09 | 0.65 | 0.78 | ✓ Certified | Batson & Royer, ICML 2019 |
| 9 | VST-Denoise + gradient | 0.520 | 20.16 | 0.606 | 0.88 | ✓ Certified | Anscombe, Biometrika 1948 |
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%