Hidden

Ocean Acoustic Tomography — 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
sound_speed_profile_error -0.28 – 0.92 -
multipath_identification -2.8 – 9.2 -
source/receiver_position -1.4 – 4.6 m
current_velocity_error -0.07 – 0.23 m/s

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SwinIR + gradient 0.650 26.04 0.833 0.78 ✓ Certified Liang et al., ICCVW 2021
2 ScoreExperimental + gradient 0.635 25.0 0.802 0.82 ✓ Certified Wei et al., 2025
3 Domain-Adapted-CNN + gradient 0.633 25.53 0.818 0.75 ✓ Certified Domain adaptation CNN
4 DiffusionExperimental + gradient 0.602 23.32 0.743 0.86 ✓ Certified Zhang et al., 2024
5 PnP-RED + gradient 0.569 22.58 0.714 0.79 ✓ Certified Romano et al., IEEE TIP 2017
6 ExpFormer + gradient 0.562 22.14 0.696 0.81 ✓ Certified Experimental science transformer, 2024
7 Tikhonov + gradient 0.560 22.46 0.709 0.76 ✓ Certified Tikhonov, Doklady 1963
8 Matched Filter + gradient 0.534 21.55 0.67 0.75 ✓ Certified Optimal linear filter
9 ResUNet + gradient 0.518 20.84 0.638 0.77 ✓ Certified Residual U-Net baseline
10 PnP-ADMM + gradient 0.505 19.7 0.584 0.87 ✓ Certified ADMM + denoiser prior
11 Wiener Filter + gradient 0.489 20.04 0.601 0.74 ✓ Certified Wiener filtering baseline

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 Ocean Acoustic Tomography