Dev
CASSI — Dev Tier
(10 scenes)Blind evaluation tier — no ground truth available.
What you get
Measurements (y), ideal forward operator (H), and spec ranges only.
How to use
Apply your pipeline from the Public tier. Use consistency as self-check.
What to submit
Reconstructed signals and corrected spec. Scored server-side.
Parameter Specifications
🔒
True spec hidden — estimate parameters from spec ranges below.
| Parameter | Spec Range | Unit |
|---|---|---|
| mask_dx | 0.4 – 0.8 | px |
| mask_dy | 0.2 – 0.6 | px |
| mask_rotation | 0.05 – 0.25 | deg |
| dispersion_slope | 1.825 – 2.075 | px/band |
| dispersion_axis | 0.07 – 0.37 | deg |
Dev Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | GAP-TV + gradient | 0.576 | 19.69 | 0.699 | 1.0 | ✓ Certified | PWM benchmark |
| 2 | SSR-L + gradient | 0.545 | 19.53 | 0.626 | 1.0 | ✓ Certified | PWM benchmark (CVPR 2024) |
| 3 | PnP-HSICNN + gradient | 0.508 | 16.88 | 0.621 | 0.95 | ✓ Certified | PWM benchmark |
| 4 | MST-L + gradient | 0.472 | 17.18 | 0.55 | 0.9 | ✓ Certified | PWM benchmark |
| 5 | HDNet + gradient | 0.329 | 13.59 | 0.427 | 0.61 | ✓ Certified | PWM benchmark |
Visible Data Fields
y
H_ideal
spec_ranges
Dataset
Format: HDF5
Scenes: 10
Scoring Formula
0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)
PSNR: 40%
SSIM: 40%
Consistency: 20%