Dev

OCT — Dev Tier

(3 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
dispersion -240.0 – 360.0 fs²
reference_delay -6.0 – 9.0 μm
spectral_roll_off -1.2 – 1.8 dB/mm

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SpeckleFormer + gradient 0.783 33.26 0.955 0.84 ✓ Certified Devalla et al., ECCV 2024
2 OCT-ViT + gradient 0.768 31.66 0.939 0.88 ✓ Certified Tian et al., ICCV 2024
3 RetinalFormer + gradient 0.765 31.56 0.938 0.87 ✓ Certified Chen et al., ICCV 2024
4 ScoreOCT + gradient 0.759 30.92 0.93 0.89 ✓ Certified Wei et al., ECCV 2025
5 DiffusionOCT + gradient 0.719 28.61 0.893 0.88 ✓ Certified Zhang et al., NeurIPS 2024
6 Speckle-Lee + gradient 0.654 25.24 0.81 0.89 ✓ Certified Lee, IEEE TGRS 1980
7 OCTA-Net + gradient 0.645 25.08 0.805 0.86 ✓ Certified Hybrid U-Net+Transformer, 2023
8 Speckle-DenoiseNet + gradient 0.637 25.37 0.814 0.79 ✓ Certified Devalla et al., BOE 2019
9 TV-Denoising + gradient 0.633 25.0 0.802 0.81 ✓ Certified Rudin et al., Phys. A 1992
10 NLM-OCT + gradient 0.632 24.68 0.792 0.84 ✓ Certified Buades et al., Multiscale Model. Simul. 2005
11 BM4D + gradient 0.629 24.39 0.782 0.86 ✓ Certified Maggioni et al., IEEE TIP 2013
12 U-Net-OCT + gradient 0.619 24.32 0.78 0.82 ✓ Certified Ronneberger et al., MICCAI 2015 (OCT variant)
13 FFT-OCT + gradient 0.612 23.76 0.76 0.85 ✓ Certified Analytical baseline

Visible Data Fields

y H_ideal spec_ranges

Dataset

Format: HDF5
Scenes: 3

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to OCT