Hidden

CDI — 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
support -2.1 – 6.9 pixels
saturation -3.5 – 11.5 %
missing_center -2.1 – 6.9 pixels

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 HolographyViT + gradient 0.702 28.4 0.889 0.81 ✓ Certified Wang et al., ICCV 2024
2 AutoPhase++ + gradient 0.691 28.35 0.888 0.76 ✓ Certified Rivenson et al., ECCV 2024
3 PhaseFormer + gradient 0.653 26.19 0.837 0.78 ✓ Certified Tian et al., ICCV 2024
4 LRGS + gradient 0.641 25.65 0.822 0.78 ✓ Certified Choi et al., 2023
5 DiffusionPhase + gradient 0.627 24.92 0.8 0.79 ✓ Certified Song et al., NeurIPS 2024
6 ScorePhase + gradient 0.590 22.81 0.724 0.86 ✓ Certified Wei et al., ECCV 2025
7 PhaseResNet + gradient 0.586 23.45 0.748 0.76 ✓ Certified Baoqing et al., Optica 2023
8 CyclePhase + gradient 0.522 20.79 0.636 0.8 ✓ Certified Ge et al., IEEE Photonics 2023
9 PhaseNet + gradient 0.504 19.68 0.583 0.87 ✓ Certified Rivenson et al., LSA 2018
10 GS/HIO + gradient 0.494 20.09 0.603 0.76 ✓ Certified Fienup, Appl. Opt. 1982
11 prDeep + gradient 0.479 19.05 0.552 0.84 ✓ Certified Metzler et al., ICML 2018
12 Error Reduction + gradient 0.463 19.03 0.551 0.76 ✓ Certified Fienup, J. Opt. Soc. Am. 1982
13 Gerchberg-Saxton + gradient 0.384 15.7 0.387 0.86 ✓ Certified Gerchberg & Saxton, Optik 1972
14 deep-PR + gradient 0.382 15.76 0.39 0.84 ✓ Certified Asif et al., ICCP 2017

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 CDI