Public
Atom Probe Tomography (APT) — 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 |
|---|---|---|---|
| flight_path_error | -0.1 – 0.2 | 0.05 | mm |
| voltage_calibration | 0.996 – 1.008 | 1.002 | - |
| detection_efficiency | 0.58 – 0.64 | 0.61 | - |
| tip_radius_error | -1.0 – 2.0 | 0.5 | nm |
InverseNet Baseline Scores
Method: CPU_baseline — Mismatch parameter: nominal
Scenario I (Ideal)
33.04 dB
SSIM 0.8313
Scenario II (Mismatch)
21.66 dB
SSIM 0.3376
Scenario III (Oracle)
22.75 dB
SSIM 0.4197
Per-scene breakdown (4 scenes)
| Scene | PSNR I | SSIM I | PSNR II | SSIM II | PSNR III | SSIM III |
|---|---|---|---|---|---|---|
| scene_00 | 32.90 | 0.8330 | 21.36 | 0.3538 | 22.69 | 0.4393 |
| scene_01 | 32.93 | 0.8361 | 21.23 | 0.3703 | 22.66 | 0.4596 |
| scene_02 | 33.03 | 0.8183 | 21.98 | 0.3053 | 22.85 | 0.3832 |
| scene_03 | 33.32 | 0.8377 | 22.06 | 0.3211 | 22.79 | 0.3966 |
Public Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | DiffusionAPT + gradient | 0.807 | 33.75 | 0.959 | 0.93 | ✓ Certified | Inspired by Chung et al., ICLR 2023 (score-based MRI) |
| 2 | EquivAPT + gradient | 0.799 | 33.42 | 0.956 | 0.91 | ✓ Certified | Adapted from equivariant vision transformer for atomic imaging, 2025 |
| 3 | APT-Former + gradient | 0.759 | 30.67 | 0.926 | 0.91 | ✓ Certified | Moody et al., Microsc. Microanal. 30(2):341, 2024 |
| 4 | TrajectoryPINN + gradient | 0.726 | 28.88 | 0.898 | 0.89 | ✓ Certified | De Geuser & Gault, Annu. Rev. Mater. Res. 52:1, 2022 |
| 5 | LISTA-APT + gradient | 0.694 | 27.23 | 0.864 | 0.88 | ✓ Certified | Gregor & LeCun, ICML 2010; adapted for APT 2020 |
| 6 | ResNet-ArtefactCorr + gradient | 0.682 | 26.77 | 0.852 | 0.87 | ✓ Certified | Wei et al., Ultramicroscopy 206:112817, 2019 |
| 7 | PnP-BM3D (APT) + gradient | 0.624 | 24.1 | 0.772 | 0.87 | ✓ Certified | Danielyan et al., IEEE TIP 21(9):3884, 2012 |
| 8 | Tikhonov-Trajectory + gradient | 0.547 | 21.02 | 0.647 | 0.89 | ✓ Certified | Geiser et al., Microsc. Microanal. 13(6):437, 2007 |
| 9 | Bas-Protocol + gradient | 0.514 | 19.67 | 0.583 | 0.92 | ✓ Certified | Bas et al., Appl. Surf. Sci. 87-88:298, 1995 |
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%