Optical Diffraction Tomography (ODT)

Optical Diffraction Tomography (ODT)

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
🥇 Rytov-Former 0.784 34.0 0.935 ✓ Certified ODT reconstruction transformer, 2024
🥈 ODT-Net 0.736 32.0 0.905 ✓ Certified Zhou et al., Light: S&A 2023
🥉 Born-ADMM 0.622 28.0 0.810 ✓ Certified Lim et al., PRL 2015
4 Wolf FBP 0.503 24.5 0.690 ✓ Certified Wolf, Opt. Commun. 1969

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
🥇 Rytov-Former + gradient 0.706
0.766
31.28 dB / 0.934
0.710
28.21 dB / 0.885
0.641
25.89 dB / 0.829
✓ Certified ODT reconstruction transformer, 2024
🥈 ODT-Net + gradient 0.663
0.742
30.07 dB / 0.918
0.650
25.07 dB / 0.804
0.598
23.22 dB / 0.740
✓ Certified Zhou et al., Light: S&A 2023
🥉 Born-ADMM + gradient 0.615
0.661
25.55 dB / 0.819
0.612
23.51 dB / 0.751
0.571
22.85 dB / 0.725
✓ Certified Lim et al., Phys. Rev. Lett. 2015
4 Wolf FBP + gradient 0.542
0.570
21.77 dB / 0.680
0.537
21.3 dB / 0.659
0.518
20.49 dB / 0.622
✓ Certified Wolf, Opt. Commun. 1969

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 5 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 Rytov-Former + gradient 0.766 31.28 0.934
2 ODT-Net + gradient 0.742 30.07 0.918
3 Born-ADMM + gradient 0.661 25.55 0.819
4 Wolf FBP + gradient 0.570 21.77 0.68
Spec Ranges (4 parameters)
Parameter Min Max Unit
illumination_angle_error -0.4 0.8 degperangle
missing_cone_artifact 26.0 38.0 deg
refractive_index_of_medium 1.3344 1.3422 -
multiple_scattering -2.0 4.0 -
View Details →
Dev 5 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 Rytov-Former + gradient 0.710 28.21 0.885
2 ODT-Net + gradient 0.650 25.07 0.804
3 Born-ADMM + gradient 0.612 23.51 0.751
4 Wolf FBP + gradient 0.537 21.3 0.659
Spec Ranges (4 parameters)
Parameter Min Max Unit
illumination_angle_error -0.48 0.72 degperangle
missing_cone_artifact 25.2 37.2 deg
refractive_index_of_medium 1.33388 1.34168 -
multiple_scattering -2.4 3.6 -
Submit Reconstruction View Details →
Hidden 5 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 Rytov-Former + gradient 0.641 25.89 0.829
2 ODT-Net + gradient 0.598 23.22 0.74
3 Born-ADMM + gradient 0.571 22.85 0.725
4 Wolf FBP + gradient 0.518 20.49 0.622
Spec Ranges (4 parameters)
Parameter Min Max Unit
illumination_angle_error -0.28 0.92 degperangle
missing_cone_artifact 27.2 39.2 deg
refractive_index_of_medium 1.33518 1.34298 -
multiple_scattering -1.4 4.6 -
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

P → D

P Propagation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
i_a illumination_angle_error Illumination angle error (deg per angle) 0.0 0.4
m_c missing_cone_artifact Missing cone artifact (deg) 30.0 34.0
r_i refractive_index_of_medium Refractive index of medium (-) 1.337 1.3396
m_s multiple_scattering Multiple scattering (-) 0.0 2.0

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.