Hidden

Polarimetric SAR (PolSAR) — Hidden Tier

(3 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.

Parameter	Spec Range	Unit
cross_talk_between_polarizations	-11.5 – 3.5	dB
channel_imbalance	-0.14 – 0.46	dB
faraday_rotation	-0.7 – 2.3	deg

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	SARFormer + gradient	0.703	27.83	0.877	0.87	✓ Certified	Li et al., CVPR 2024
2	PanSharpener++ + gradient	0.653	25.47	0.817	0.86	✓ Certified	Zhang et al., ICCV 2024
3	DiffusionSAR + gradient	0.638	25.79	0.826	0.75	✓ Certified	Wei et al., NeurIPS 2024
4	ScoreSAR + gradient	0.630	24.67	0.792	0.83	✓ Certified	Johnson et al., ECCV 2025
5	SARDenoiserViT + gradient	0.620	24.62	0.79	0.79	✓ Certified	Wang et al., ICCV 2024
6	Range-Doppler + gradient	0.588	23.46	0.749	0.77	✓ Certified	SAR signal processing baseline
7	SAR-CAM + gradient	0.578	23.08	0.734	0.77	✓ Certified	Cross-attention SAR, 2024
8	SAR-DRN + gradient	0.566	22.82	0.724	0.74	✓ Certified	Zhang et al., RS 2018
9	Chirp Scaling + gradient	0.557	21.51	0.669	0.87	✓ Certified	Raney et al., IEEE TGRS 1994
10	SAR-BM3D + gradient	0.531	21.3	0.659	0.77	✓ Certified	Parrilli et al., IEEE TGRS 2012
11	Lee Filter + gradient	0.515	20.89	0.641	0.75	✓ Certified	Lee, IEEE TGRS 1980
12	Matched Filter + gradient	0.458	18.67	0.533	0.79	✓ Certified	Standard SAR focusing
13	SAR-ResNet + gradient	0.420	17.68	0.484	0.75	✓ Certified	Chen et al., IEEE TGRS 2022

Scenes: 3

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

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