Public

SEM — 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
beam_energy -0.1 – 0.2 0.05 keV
stigmatism -5.0 – 10.0 2.5 nm
working_distance -0.1 – 0.2 0.05 mm

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

28.55 dB

SSIM 0.8588

Scenario II (Mismatch)

12.54 dB

SSIM 0.2332

Scenario III (Oracle)

12.09 dB

SSIM 0.0486

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 27.21 0.9294 9.88 0.1554 11.93 0.0716
scene_01 28.20 0.9225 7.15 0.0915 9.12 0.0495
scene_02 35.42 0.7281 18.08 0.5046 13.45 0.0095
scene_03 23.39 0.8550 15.03 0.1812 13.86 0.0636

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SwinIR + gradient 0.763 31.55 0.938 0.86 ✓ Certified Liang et al., ICCVW 2021
2 Noise2Void + gradient 0.735 29.74 0.913 0.86 ✓ Certified Krull et al., CVPR 2019
3 BM3D + gradient 0.669 25.73 0.824 0.91 ✓ Certified Dabov et al., IEEE TIP 2007
4 Wiener Filter + gradient 0.618 23.39 0.746 0.93 ✓ Certified Analytical baseline

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 SEM