Public

PALM/STORM — Public Tier

(5 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
psf_model -5.0 – 10.0 2.5 GaussianvsAiry
emitter_density -20.0 – 40.0 10.0 %
drift -0.5 – 1.0 0.25 nm/frame

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

9.86 dB

SSIM 0.1684

Scenario II (Mismatch)

9.89 dB

SSIM 0.1626

Scenario III (Oracle)

13.03 dB

SSIM 0.3455

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 11.37 0.1703 11.55 0.1716 14.78 0.3748
scene_01 11.35 0.2400 11.40 0.2326 14.49 0.4899
scene_02 8.13 0.1298 8.02 0.1186 11.08 0.2512
scene_03 8.57 0.1336 8.60 0.1278 11.76 0.2659

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DECODE + gradient 0.741 30.01 0.917 0.87 ✓ Certified Speiser et al., Nat. Methods 2021
2 Deep-STORM + gradient 0.737 29.17 0.903 0.92 ✓ Certified Nehme et al., Optica 2018
3 FALCON + gradient 0.641 24.21 0.776 0.94 ✓ Certified Min et al., Sci. Rep. 2014
4 ThunderSTORM + gradient 0.522 20.24 0.61 0.88 ✓ Certified Ovesny et al., Bioinformatics 2014

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 5

Scoring Formula

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

PSNR: 40% SSIM: 40% Consistency: 20%
Back to PALM/STORM