Public

SAR — Public Tier

(3 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

✓

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
motion_error -2.0 – 4.0 1.0 cm
phase_error -0.3 – 0.6 0.15 rad
range_cell_migration -0.5 – 1.0 0.25 cells

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

25.50 dB

SSIM 0.7080

Scenario II (Mismatch)

25.49 dB

SSIM 0.7034

Scenario III (Oracle)

27.77 dB

SSIM 0.7333

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 15.50 0.2342 15.45 0.2374 15.55 0.2305
scene_01 29.03 0.8684 28.59 0.8712 31.59 0.9076
scene_02 29.23 0.8516 28.33 0.8571 31.37 0.8998
scene_03 28.23 0.8777 29.58 0.8479 32.56 0.8955

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PanSharpener++ + gradient 0.803 33.58 0.958 0.92 ✓ Certified Zhang et al., ICCV 2024
2 DiffusionSAR + gradient 0.789 33.01 0.953 0.89 ✓ Certified Wei et al., NeurIPS 2024
3 SARFormer + gradient 0.762 30.88 0.929 0.91 ✓ Certified Li et al., CVPR 2024
4 SAR-CAM + gradient 0.738 29.47 0.908 0.9 ✓ Certified Cross-attention SAR, 2024
5 ScoreSAR + gradient 0.732 28.9 0.898 0.92 ✓ Certified Johnson et al., ECCV 2025
6 SARDenoiserViT + gradient 0.732 28.51 0.891 0.95 ✓ Certified Wang et al., ICCV 2024
7 SAR-DRN + gradient 0.717 28.46 0.89 0.88 ✓ Certified Zhang et al., RS 2018
8 SAR-ResNet + gradient 0.708 27.36 0.867 0.94 ✓ Certified Chen et al., IEEE TGRS 2022
9 Lee Filter + gradient 0.705 27.18 0.862 0.94 ✓ Certified Lee, IEEE TGRS 1980
10 SAR-BM3D + gradient 0.645 24.9 0.799 0.88 ✓ Certified Parrilli et al., IEEE TGRS 2012
11 Range-Doppler + gradient 0.617 23.74 0.759 0.88 ✓ Certified SAR signal processing baseline
12 Chirp Scaling + gradient 0.558 21.01 0.646 0.95 ✓ Certified Raney et al., IEEE TGRS 1994
13 Matched Filter + gradient 0.542 20.79 0.636 0.9 ✓ Certified Standard SAR focusing

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 3

Scoring Formula

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

PSNR: 40% SSIM: 40% Consistency: 20%
Back to SAR