Dev

Seismic Tomography — Dev Tier

(5 scenes)

Blind evaluation tier — no ground truth available.

Measurements (y), ideal forward operator (H), and spec ranges only.

Apply your pipeline from the Public tier. Use consistency as self-check.

Reconstructed signals and corrected spec. Scored server-side.

🔒

True spec hidden — estimate parameters from spec ranges below.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	ExpFormer + gradient	0.709	28.74	0.895	0.82	✓ Certified	Experimental science transformer, 2024
2	SwinIR + gradient	0.705	28.26	0.886	0.84	✓ Certified	Liang et al., ICCVW 2021
3	ScoreExperimental + gradient	0.655	25.65	0.822	0.85	✓ Certified	Wei et al., 2025
4	ResUNet + gradient	0.654	25.39	0.814	0.87	✓ Certified	Residual U-Net baseline
5	Domain-Adapted-CNN + gradient	0.626	24.86	0.798	0.79	✓ Certified	Domain adaptation CNN
6	PnP-RED + gradient	0.623	24.76	0.794	0.79	✓ Certified	Romano et al., IEEE TIP 2017
7	DiffusionExperimental + gradient	0.615	24.38	0.782	0.79	✓ Certified	Zhang et al., 2024
8	Tikhonov + gradient	0.579	22.69	0.719	0.82	✓ Certified	Tikhonov, Doklady 1963
9	Matched Filter + gradient	0.578	22.13	0.696	0.89	✓ Certified	Optimal linear filter
10	Wiener Filter + gradient	0.570	21.94	0.687	0.88	✓ Certified	Wiener filtering baseline
11	PnP-ADMM + gradient	0.526	21.05	0.648	0.78	✓ Certified	ADMM + denoiser prior

y H_ideal spec_ranges

Format: HDF5

Scenes: 5

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

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