Ocean Color Remote Sensing

Ocean Color Remote Sensing

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

# Method Score PSNR (dB) SSIM Source
🥇 AquaFormer 0.747 32.5 0.910 ✓ Certified Ocean color transformer, 2024
🥈 OC-Net 0.693 30.5 0.870 ✓ Certified Pahlevan et al., RSE 2022
🥉 MUMM 0.553 26.0 0.740 ✓ Certified Ruddick et al., RSE 2000
4 Gordon AC 0.430 22.5 0.610 ✓ Certified Gordon & Wang, Appl. Opt. 1994

Dataset: PWM Benchmark (4 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

# Method Overall Score Public
PSNR / SSIM
 Dev
PSNR / SSIM
 Hidden
PSNR / SSIM
 Trust Source
🥇 AquaFormer + gradient 0.712
0.751
30.61 dB / 0.926
0.713
28.17 dB / 0.884
0.673
26.23 dB / 0.838
✓ Certified Ocean color retrieval transformer, 2024
🥈 OC-Net + gradient 0.624
0.709
27.8 dB / 0.877
0.608
23.38 dB / 0.746
0.556
22.44 dB / 0.708
✓ Certified Pahlevan et al., RSE 2022
🥉 MUMM + gradient 0.593
0.612
23.37 dB / 0.745
0.589
22.71 dB / 0.720
0.578
22.81 dB / 0.724
✓ Certified Ruddick et al., RSE 2000
4 Gordon AC + gradient 0.474
0.557
21.03 dB / 0.647
0.453
18.03 dB / 0.501
0.411
16.84 dB / 0.442
✓ Certified Gordon & Wang, Appl. Opt. 1994

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →
Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%
Public 3 scenes

Full-access development tier with all data visible.

What you get & how to use

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.

Public Leaderboard
# Method Score PSNR SSIM
1 AquaFormer + gradient 0.751 30.61 0.926
2 OC-Net + gradient 0.709 27.8 0.877
3 MUMM + gradient 0.612 23.37 0.745
4 Gordon AC + gradient 0.557 21.03 0.647
Spec Ranges (3 parameters)
Parameter Min Max Unit
atmospheric_correction_error -3.0 6.0 -
sun_glint_contamination -4.0 8.0 -
vicarious_calibration_offset -0.6 1.2 -
View Details →
Dev 3 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

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.

Dev Leaderboard
# Method Score PSNR SSIM
1 AquaFormer + gradient 0.713 28.17 0.884
2 OC-Net + gradient 0.608 23.38 0.746
3 MUMM + gradient 0.589 22.71 0.72
4 Gordon AC + gradient 0.453 18.03 0.501
Spec Ranges (3 parameters)
Parameter Min Max Unit
atmospheric_correction_error -3.6 5.4 -
sun_glint_contamination -4.8 7.2 -
vicarious_calibration_offset -0.72 1.08 -
Submit Reconstruction View Details →
Hidden 3 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

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.

Hidden Leaderboard
# Method Score PSNR SSIM
1 AquaFormer + gradient 0.673 26.23 0.838
2 MUMM + gradient 0.578 22.81 0.724
3 OC-Net + gradient 0.556 22.44 0.708
4 Gordon AC + gradient 0.411 16.84 0.442
Spec Ranges (3 parameters)
Parameter Min Max Unit
atmospheric_correction_error -2.1 6.9 -
sun_glint_contamination -2.8 9.2 -
vicarious_calibration_offset -0.42 1.38 -
Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

M → Σ → D

M Modulation
Σ Summation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
a_c atmospheric_correction_error Atmospheric correction error (-) 0.0 3.0
s_g sun_glint_contamination Sun glint contamination (-) 0.0 4.0
v_c vicarious_calibration_offset Vicarious calibration offset (-) 0.0 0.6

Credits System

40%
Platform Profit Pool
Revenue allocated to benchmark rewards
30%
Winner Share
Top algorithm receives from pool
$100
Min Withdrawal
Minimum payout threshold
Spec Primitives Reference (11 primitives)
P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.