Ultrasonic Phased Array (TFM/FMC)

Ultrasonic Phased Array (TFM/FMC)

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
🥇 FMC-Former 0.795 34.5 0.940 ✓ Certified Full matrix capture transformer, 2024
🥈 UTPA-Net 0.744 32.5 0.905 ✓ Certified Phased array DL, 2022
🥉 SAFT 0.622 28.0 0.810 ✓ Certified Doctor et al., 1986
4 TFM 0.522 25.0 0.710 ✓ Certified Holmes et al., 2005

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
🥇 FMC-Former + gradient 0.719
0.777
32.43 dB / 0.947
0.714
29.09 dB / 0.902
0.665
26.6 dB / 0.848
✓ Certified Full matrix capture transformer, 2024
🥈 UTPA-Net + gradient 0.667
0.771
31.14 dB / 0.933
0.632
24.28 dB / 0.778
0.598
23.54 dB / 0.752
✓ Certified Phased array DL reconstruction, 2022
🥉 SAFT + gradient 0.633
0.692
26.54 dB / 0.847
0.635
24.33 dB / 0.780
0.572
22.24 dB / 0.700
✓ Certified Doctor et al., NDT Int. 1986
4 TFM + gradient 0.589
0.628
23.79 dB / 0.761
0.592
23.0 dB / 0.731
0.547
21.62 dB / 0.673
✓ Certified Holmes et al., NDT&E Int. 2005

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 FMC-Former + gradient 0.777 32.43 0.947
2 UTPA-Net + gradient 0.771 31.14 0.933
3 SAFT + gradient 0.692 26.54 0.847
4 TFM + gradient 0.628 23.79 0.761
Spec Ranges (4 parameters)
Parameter Min Max Unit
element_pitch_error -0.002 0.004 mm
sound_velocity 5860.0 5980.0 m/s
wedge_angle_error -0.2 0.4 deg
dead_element_fraction -0.01 0.02 -
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 FMC-Former + gradient 0.714 29.09 0.902
2 SAFT + gradient 0.635 24.33 0.78
3 UTPA-Net + gradient 0.632 24.28 0.778
4 TFM + gradient 0.592 23.0 0.731
Spec Ranges (4 parameters)
Parameter Min Max Unit
element_pitch_error -0.0024 0.0036 mm
sound_velocity 5852.0 5972.0 m/s
wedge_angle_error -0.24 0.36 deg
dead_element_fraction -0.012 0.018 -
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 FMC-Former + gradient 0.665 26.6 0.848
2 UTPA-Net + gradient 0.598 23.54 0.752
3 SAFT + gradient 0.572 22.24 0.7
4 TFM + gradient 0.547 21.62 0.673
Spec Ranges (4 parameters)
Parameter Min Max Unit
element_pitch_error -0.0014 0.0046 mm
sound_velocity 5872.0 5992.0 m/s
wedge_angle_error -0.14 0.46 deg
dead_element_fraction -0.007 0.023 -
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
e_p element_pitch_error Element pitch error (mm) 0.0 0.002
s_v sound_velocity Sound velocity (m/s) 5900.0 5940.0
w_a wedge_angle_error Wedge angle error (deg) 0.0 0.2
d_e dead_element_fraction Dead element fraction (-) 0.0 0.01

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.