Hidden

Confocal Live-Cell — 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
pinhole -3.5 – 11.5 μm
refractive_index 1.5115 – 1.5265
photobleaching -3.5 – 11.5 %

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DiffusionCell + gradient 0.694 28.42 0.889 0.77 ✓ Certified Gao et al., Nat. Methods 2024
2 Restormer-Micro + gradient 0.678 26.52 0.846 0.87 ✓ Certified Zamir et al., CVPR 2022 (microscopy)
3 SwinIR-LiveCell + gradient 0.669 26.67 0.85 0.81 ✓ Certified Liang et al., ICCV 2021 (live-cell)
4 Noise2Void + gradient 0.575 22.26 0.701 0.86 ✓ Certified Krull et al., CVPR 2019
5 PN2V + gradient 0.574 22.67 0.718 0.8 ✓ Certified Krull et al., ECCV 2020
6 CARE + gradient 0.547 21.55 0.67 0.82 ✓ Certified Weigert et al., Nat. Methods 2018
7 NLM-Fluorescence + gradient 0.545 21.02 0.647 0.88 ✓ Certified Buades et al., CVPR 2005
8 Noise2Self + gradient 0.486 19.4 0.57 0.82 ✓ Certified Batson & Royer, ICML 2019
9 VST-Denoise + gradient 0.478 19.34 0.567 0.79 ✓ Certified Anscombe, Biometrika 1948

Dataset

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 Confocal Live-Cell