Public
LiDAR — 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 |
|---|---|---|---|
| timing_jitter | -50.0 – 100.0 | 25.0 | ps |
| beam_divergence | -0.1 – 0.2 | 0.05 | mrad |
| range_walk | -1.0 – 2.0 | 0.5 | cm |
InverseNet Baseline Scores
Method: CPU_baseline — Mismatch parameter: nominal
Scenario I (Ideal)
7.41 dB
SSIM 0.4486
Scenario II (Mismatch)
7.10 dB
SSIM 0.2412
Scenario III (Oracle)
17.25 dB
SSIM 0.2762
Per-scene breakdown (4 scenes)
| Scene | PSNR I | SSIM I | PSNR II | SSIM II | PSNR III | SSIM III |
|---|---|---|---|---|---|---|
| scene_00 | 12.80 | 0.3797 | 12.07 | 0.3583 | 13.51 | 0.4022 |
| scene_01 | 5.81 | 0.4796 | 6.03 | 0.1934 | 18.47 | 0.2267 |
| scene_02 | 5.66 | 0.4660 | 5.08 | 0.2092 | 18.61 | 0.2372 |
| scene_03 | 5.39 | 0.4690 | 5.23 | 0.2041 | 18.40 | 0.2387 |
Public Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | Point Transformer + gradient | 0.758 | 31.09 | 0.932 | 0.87 | ✓ Certified | Zhao et al., ICCV 2021 |
| 2 | RandLA-Net + gradient | 0.740 | 30.07 | 0.918 | 0.86 | ✓ Certified | Hu et al., CVPR 2020 |
| 3 | PnP-ADMM + gradient | 0.679 | 26.23 | 0.838 | 0.91 | ✓ Certified | Venkatakrishnan et al., 2013 |
| 4 | Bilateral Filter + gradient | 0.650 | 25.05 | 0.804 | 0.89 | ✓ Certified | Tomasi & Manduchi, ICCV 1998 |
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%