Hidden

Seismic Tomography — Hidden Tier

(5 scenes)

Fully blind server-side evaluation — no data download.

No data downloadable. Algorithm runs server-side on hidden measurements.

Package algorithm as Docker container / Python script. Submit via link.

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

🔒

True spec hidden — blind evaluation, only ranges available.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	ExpFormer + gradient	0.651	26.23	0.838	0.77	✓ Certified	Experimental science transformer, 2024
2	ResUNet + gradient	0.606	23.59	0.754	0.84	✓ Certified	Residual U-Net baseline
3	SwinIR + gradient	0.605	24.05	0.77	0.78	✓ Certified	Liang et al., ICCVW 2021
4	ScoreExperimental + gradient	0.601	24.24	0.777	0.74	✓ Certified	Wei et al., 2025
5	PnP-RED + gradient	0.588	23.39	0.746	0.78	✓ Certified	Romano et al., IEEE TIP 2017
6	Domain-Adapted-CNN + gradient	0.567	22.09	0.694	0.84	✓ Certified	Domain adaptation CNN
7	Wiener Filter + gradient	0.553	21.89	0.685	0.8	✓ Certified	Wiener filtering baseline
8	DiffusionExperimental + gradient	0.548	21.87	0.684	0.78	✓ Certified	Zhang et al., 2024
9	Matched Filter + gradient	0.547	21.75	0.679	0.79	✓ Certified	Optimal linear filter
10	Tikhonov + gradient	0.527	21.03	0.647	0.79	✓ Certified	Tikhonov, Doklady 1963
11	PnP-ADMM + gradient	0.434	17.6	0.48	0.83	✓ Certified	ADMM + denoiser prior

Scenes: 5

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%