Hyperspectral Remote Sensing

Hyperspectral 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
🥇 MST++ 0.841 36.8 0.955 ✓ Certified Cai et al., CVPRW 2022
🥈 DBIN 0.790 34.5 0.930 ✓ Certified Dong et al., CVPR 2021
🥉 PnP-LTTR 0.675 30.0 0.850 ✓ Certified He et al., IEEE TGRS 2020
4 CNMF 0.543 26.0 0.720 ✓ Certified Yokoya et al., IEEE TGRS 2012

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
🥇 MST++ + gradient 0.738
0.828
35.21 dB / 0.969
0.729
29.16 dB / 0.903
0.657
26.18 dB / 0.837
✓ Certified Cai et al., CVPRW 2022
🥈 DBIN + gradient 0.643
0.774
31.76 dB / 0.940
0.629
24.76 dB / 0.794
0.525
20.62 dB / 0.628
✓ Certified Dong et al., CVPR 2021
🥉 CNMF + gradient 0.607
0.623
24.06 dB / 0.771
0.613
24.15 dB / 0.774
0.586
23.4 dB / 0.746
✓ Certified Yokoya et al., IEEE TGRS 2012
4 PnP-LTTR + gradient 0.601
0.699
27.17 dB / 0.862
0.583
22.48 dB / 0.710
0.521
20.25 dB / 0.611
✓ Certified He et al., IEEE TGRS 2020

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 MST++ + gradient 0.828 35.21 0.969
2 DBIN + gradient 0.774 31.76 0.94
3 PnP-LTTR + gradient 0.699 27.17 0.862
4 CNMF + gradient 0.623 24.06 0.771
Spec Ranges (4 parameters)
Parameter Min Max Unit
spectral_shift -0.4 0.8 nm
smile_distortion -0.2 0.4 px
keystone_distortion -0.1 0.2 px
radiometric_gain 0.98 1.04 -
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 MST++ + gradient 0.729 29.16 0.903
2 DBIN + gradient 0.629 24.76 0.794
3 CNMF + gradient 0.613 24.15 0.774
4 PnP-LTTR + gradient 0.583 22.48 0.71
Spec Ranges (4 parameters)
Parameter Min Max Unit
spectral_shift -0.48 0.72 nm
smile_distortion -0.24 0.36 px
keystone_distortion -0.12 0.18 px
radiometric_gain 0.976 1.036 -
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 MST++ + gradient 0.657 26.18 0.837
2 CNMF + gradient 0.586 23.4 0.746
3 DBIN + gradient 0.525 20.62 0.628
4 PnP-LTTR + gradient 0.521 20.25 0.611
Spec Ranges (4 parameters)
Parameter Min Max Unit
spectral_shift -0.28 0.92 nm
smile_distortion -0.14 0.46 px
keystone_distortion -0.07 0.23 px
radiometric_gain 0.986 1.046 -
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 → W → Σ → D

M Modulation
W Warp
Σ Summation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
s_s spectral_shift Spectral shift (nm) 0.0 0.4
s_d smile_distortion Smile distortion (px) 0.0 0.2
k_d keystone_distortion Keystone distortion (px) 0.0 0.1
r_g radiometric_gain Radiometric gain (-) 1.0 1.02

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.