Hidden

Matrix — 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
matrix_perturb -0.007 – 0.023
gain 0.979 – 1.069
sigma_y -0.014 – 0.046

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 ScoreSCI + gradient 0.732 30.43 0.923 0.79 ✓ Certified Chen et al., NeurIPS 2024
2 FlowHSI + gradient 0.694 28.56 0.892 0.76 ✓ Certified Huang et al., arXiv 2025
3 CST + gradient 0.693 27.92 0.879 0.81 ✓ Certified Liu et al., ICCV 2023
4 PromptSCI + gradient 0.692 28.44 0.89 0.76 ✓ Certified Bai et al., ICCV 2024
5 MST-L + gradient 0.689 27.4 0.868 0.84 ✓ Certified Cai et al., CVPR 2022
6 CSTrans + gradient 0.683 26.99 0.858 0.85 ✓ Certified Liu et al., CVPR 2024
7 Restormer + gradient 0.676 27.57 0.871 0.76 ✓ Certified Zamir et al., CVPR 2022
8 HiSViT+ + gradient 0.657 25.61 0.821 0.86 ✓ Certified Tao et al., ECCV 2024
9 EfficientSCI + gradient 0.655 26.19 0.837 0.79 ✓ Certified Wang et al., IEEE TIP 2023
10 DiffusionHSI + gradient 0.646 25.32 0.812 0.84 ✓ Certified Zhang et al., ICCV 2024
11 TVAL3 + gradient 0.623 24.99 0.802 0.76 ✓ Certified Li et al., SIAM J. Sci. Comput. 2009
12 FISTA-TV + gradient 0.600 24.03 0.77 0.76 ✓ Certified Beck & Teboulle, SIAM J. Imaging Sci. 2009
13 GAP-TV + gradient 0.567 22.64 0.717 0.77 ✓ Certified Yuan et al., IEEE TIP 2016
14 PnP-FFDNet + gradient 0.525 21.18 0.654 0.76 ✓ Certified Zhang et al., IEEE TPAMI 2020

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 Matrix