Hidden
Panorama — Hidden Tier
(5 scenes)Fully blind server-side evaluation — no data download.
What you get
No data downloadable. Algorithm runs server-side on hidden measurements.
How to use
Package algorithm as Docker container / Python script. Submit via link.
What to submit
Containerized algorithm accepting y + H, outputting x_hat + corrected spec.
Parameter Specifications
🔒
True spec hidden — blind evaluation, only ranges available.
| Parameter | Spec Range | Unit |
|---|---|---|
| focus_step | -1.4 – 4.6 | μm |
| registration | -0.35 – 1.15 | pixels |
| exposure_variation | -2.1 – 6.9 | % |
Hidden Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | PanoFormer + gradient | 0.671 | 26.89 | 0.855 | 0.8 | ✓ Certified | Image stitching transformer, 2024 |
| 2 | APAP + gradient | 0.614 | 23.71 | 0.758 | 0.87 | ✓ Certified | Zaragoza et al., CVPR 2013 |
| 3 | UDIS + gradient | 0.594 | 23.86 | 0.764 | 0.75 | ✓ Certified | Nie et al., ICCV 2021 |
| 4 | SIFT-RANSAC + gradient | 0.556 | 21.76 | 0.68 | 0.83 | ✓ Certified | Lowe, IJCV 2004 |
Dataset
Scenes: 5
Scoring Formula
0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)
PSNR: 40%
SSIM: 40%
Consistency: 20%