Dev
NeRF — Dev Tier
(5 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 |
|---|---|---|
| camera_pose | -1.2 – 1.8 | mm/deg |
| focal_length | -6.0 – 9.0 | pixels |
| distortion | -0.012 – 0.018 | |
| exposure | -12.0 – 18.0 | % |
Dev Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | NeRFactor2 + gradient | 0.735 | 30.14 | 0.919 | 0.83 | ✓ Certified | Barron et al., NeurIPS 2024 |
| 2 | GaussianShader + gradient | 0.711 | 28.02 | 0.881 | 0.89 | ✓ Certified | Wang et al., ICCV 2024 |
| 3 | 3D-GS++ + gradient | 0.688 | 27.15 | 0.862 | 0.86 | ✓ Certified | Kerbl et al., SIGGRAPH 2024 |
| 4 | 3D-GS + gradient | 0.638 | 25.05 | 0.804 | 0.83 | ✓ Certified | Kerbl et al., SIGGRAPH 2023 |
| 5 | Instant-NGP + gradient | 0.623 | 23.91 | 0.765 | 0.89 | ✓ Certified | Muller et al., SIGGRAPH 2022 |
| 6 | COLMAP+MVS + gradient | 0.611 | 24.22 | 0.776 | 0.79 | ✓ Certified | Schonberger & Frahm, CVPR 2016 |
| 7 | NeRF + gradient | 0.602 | 23.31 | 0.743 | 0.86 | ✓ Certified | Mildenhall et al., ECCV 2020 |
| 8 | 2DGS + gradient | 0.598 | 23.12 | 0.736 | 0.86 | ✓ Certified | Huang et al., CVPR 2024 |
| 9 | Photogrammetry + gradient | 0.585 | 22.65 | 0.717 | 0.86 | ✓ Certified | Structure-from-Motion baseline |
| 10 | Mesh-GS + gradient | 0.512 | 20.36 | 0.616 | 0.81 | ✓ Certified | Li et al., ECCV 2024 |
| 11 | Mip-NeRF 360 + gradient | 0.497 | 19.45 | 0.572 | 0.87 | ✓ Certified | Barron et al., CVPR 2022 |
Visible Data Fields
y
H_ideal
spec_ranges
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%