Three-Photon Microscopy

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
🥇	ScoreMicro ScoreMicro Wei et al., ECCV 2025 38.48 dB SSIM 0.981 Checkpoint unavailable	0.882	38.48	0.981	✓ Certified	Wei et al., ECCV 2025
🥈	DiffDeconv DiffDeconv Huang et al., NeurIPS 2024 38.12 dB SSIM 0.979 Checkpoint unavailable	0.875	38.12	0.979	✓ Certified	Huang et al., NeurIPS 2024
🥉	Restormer+ Restormer+ Zamir et al., ICCV 2024 37.65 dB SSIM 0.975 Checkpoint unavailable	0.865	37.65	0.975	✓ Certified	Zamir et al., ICCV 2024
4	DeconvFormer DeconvFormer Chen et al., CVPR 2024 37.25 dB SSIM 0.972 Checkpoint unavailable	0.857	37.25	0.972	✓ Certified	Chen et al., CVPR 2024
5	ResUNet ResUNet DeCelle et al., Nat. Methods 2021 35.85 dB SSIM 0.964 Checkpoint unavailable	0.830	35.85	0.964	✓ Certified	DeCelle et al., Nat. Methods 2021
6	Restormer Restormer Zamir et al., CVPR 2022 35.8 dB SSIM 0.962 Checkpoint unavailable	0.828	35.8	0.962	✓ Certified	Zamir et al., CVPR 2022
7	U-Net U-Net Ronneberger et al., MICCAI 2015 35.15 dB SSIM 0.956 Checkpoint unavailable	0.814	35.15	0.956	✓ Certified	Ronneberger et al., MICCAI 2015
8	CARE CARE Weigert et al., Nat. Methods 2018 34.5 dB SSIM 0.948 Checkpoint unavailable	0.799	34.5	0.948	✓ Certified	Weigert et al., Nat. Methods 2018
9	PnP-DnCNN PnP-DnCNN Zhang et al., IEEE TIP 2017 31.2 dB SSIM 0.890 Try in SpecLab →	0.715	31.2	0.890	✓ Certified	Zhang et al., IEEE TIP 2017
10	PnP-FISTA PnP-FISTA Bai et al., 2020 30.42 dB SSIM 0.872 Try in SpecLab →	0.693	30.42	0.872	✓ Certified	Bai et al., 2020
11	TV-Deconvolution TV-Deconvolution TV-regularized deconvolution 29.5 dB SSIM 0.845 Try in SpecLab →	0.664	29.5	0.845	✓ Certified	TV-regularized deconvolution
12	Wiener Filter Wiener Filter Analytical baseline 28.35 dB SSIM 0.805 Try in SpecLab →	0.625	28.35	0.805	✓ Certified	Analytical baseline
13	Richardson-Lucy Richardson-Lucy Richardson 1972 / Lucy 1974 27.1 dB SSIM 0.770 Try in SpecLab →	0.587	27.1	0.770	✓ Certified	Richardson 1972 / Lucy 1974

Dataset: PWM Benchmark (13 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
🥇	ScoreMicro + gradient ScoreMicro + gradient Wei et al., ECCV 2025 Score 0.781 Correct & Reconstruct →	0.781	0.827 35.78 dB / 0.972	0.769 32.37 dB / 0.947	0.746 31.24 dB / 0.934	✓ Certified	Wei et al., ECCV 2025
🥈	DeconvFormer + gradient DeconvFormer + gradient Chen et al., CVPR 2024 Score 0.767 Correct & Reconstruct →	0.767	0.811 34.39 dB / 0.964	0.778 33.21 dB / 0.954	0.713 29.78 dB / 0.913	✓ Certified	Chen et al., CVPR 2024
🥉	ResUNet + gradient ResUNet + gradient DeCelle et al., Nat. Methods 2021 Score 0.743 Correct & Reconstruct →	0.743	0.792 32.95 dB / 0.952	0.732 29.2 dB / 0.904	0.705 28.69 dB / 0.895	✓ Certified	DeCelle et al., Nat. Methods 2021
4	Restormer+ + gradient Restormer+ + gradient Zamir et al., ICCV 2024 Score 0.743 Correct & Reconstruct →	0.743	0.816 35.12 dB / 0.968	0.741 29.64 dB / 0.911	0.671 26.98 dB / 0.858	✓ Certified	Zamir et al., ICCV 2024
5	DiffDeconv + gradient DiffDeconv + gradient Huang et al., NeurIPS 2024 Score 0.741 Correct & Reconstruct →	0.741	0.822 35.41 dB / 0.970	0.724 29.79 dB / 0.914	0.677 27.49 dB / 0.870	✓ Certified	Huang et al., NeurIPS 2024
6	Restormer + gradient Restormer + gradient Zamir et al., CVPR 2022 Score 0.727 Correct & Reconstruct →	0.727	0.793 33.18 dB / 0.954	0.724 29.84 dB / 0.914	0.663 27.01 dB / 0.858	✓ Certified	Zamir et al., CVPR 2022
7	CARE + gradient CARE + gradient Weigert et al., Nat. Methods 2018 Score 0.716 Correct & Reconstruct →	0.716	0.772 31.54 dB / 0.937	0.719 28.95 dB / 0.899	0.658 26.64 dB / 0.849	✓ Certified	Weigert et al., Nat. Methods 2018
8	U-Net + gradient U-Net + gradient Ronneberger et al., MICCAI 2015 Score 0.713 Correct & Reconstruct →	0.713	0.808 33.83 dB / 0.960	0.709 28.38 dB / 0.889	0.623 25.0 dB / 0.802	✓ Certified	Ronneberger et al., MICCAI 2015
9	TV-Deconvolution + gradient TV-Deconvolution + gradient Rudin et al., Phys. A 1992 Score 0.670 Correct & Reconstruct →	0.670	0.723 28.35 dB / 0.888	0.665 25.74 dB / 0.825	0.623 24.32 dB / 0.780	✓ Certified	Rudin et al., Phys. A 1992
10	PnP-DnCNN + gradient PnP-DnCNN + gradient Zhang et al., IEEE TIP 2017 Score 0.668 Correct & Reconstruct →	0.668	0.750 29.83 dB / 0.914	0.628 24.35 dB / 0.781	0.627 24.66 dB / 0.791	✓ Certified	Zhang et al., IEEE TIP 2017
11	Wiener Filter + gradient Wiener Filter + gradient Analytical baseline Score 0.645 Correct & Reconstruct →	0.645	0.664 25.48 dB / 0.817	0.646 24.88 dB / 0.798	0.625 24.85 dB / 0.797	✓ Certified	Analytical baseline
12	Richardson-Lucy + gradient Richardson-Lucy + gradient Richardson, JOSA 1972 / Lucy, AJ 1974 Score 0.608 Correct & Reconstruct →	0.608	0.648 25.13 dB / 0.806	0.594 22.76 dB / 0.722	0.581 22.7 dB / 0.719	✓ Certified	Richardson, JOSA 1972 / Lucy, AJ 1974
13	PnP-FISTA + gradient PnP-FISTA + gradient Bai et al., 2020 Score 0.608 Correct & Reconstruct →	0.608	0.704 27.48 dB / 0.869	0.594 23.61 dB / 0.754	0.525 20.47 dB / 0.621	✓ Certified	Bai et al., 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 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	ScoreMicro + gradient	0.827	35.78	0.972
2	DiffDeconv + gradient	0.822	35.41	0.97
3	Restormer+ + gradient	0.816	35.12	0.968
4	DeconvFormer + gradient	0.811	34.39	0.964
5	U-Net + gradient	0.808	33.83	0.96
6	Restormer + gradient	0.793	33.18	0.954
7	ResUNet + gradient	0.792	32.95	0.952
8	CARE + gradient	0.772	31.54	0.937
9	PnP-DnCNN + gradient	0.750	29.83	0.914
10	TV-Deconvolution + gradient	0.723	28.35	0.888
11	PnP-FISTA + gradient	0.704	27.48	0.869
12	Wiener Filter + gradient	0.664	25.48	0.817
13	Richardson-Lucy + gradient	0.648	25.13	0.806

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
scattering_coeff	10.0	25.0	mm^-1
excitation_wavelength_shift	-2.0	4.0	nm
depth_dependent_psf	-0.4	0.8	-

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	DeconvFormer + gradient	0.778	33.21	0.954
2	ScoreMicro + gradient	0.769	32.37	0.947
3	Restormer+ + gradient	0.741	29.64	0.911
4	ResUNet + gradient	0.732	29.2	0.904
5	DiffDeconv + gradient	0.724	29.79	0.914
6	Restormer + gradient	0.724	29.84	0.914
7	CARE + gradient	0.719	28.95	0.899
8	U-Net + gradient	0.709	28.38	0.889
9	TV-Deconvolution + gradient	0.665	25.74	0.825
10	Wiener Filter + gradient	0.646	24.88	0.798
11	PnP-DnCNN + gradient	0.628	24.35	0.781
12	Richardson-Lucy + gradient	0.594	22.76	0.722
13	PnP-FISTA + gradient	0.594	23.61	0.754

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
scattering_coeff	9.0	24.0	mm^-1
excitation_wavelength_shift	-2.4	3.6	nm
depth_dependent_psf	-0.48	0.72	-

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	ScoreMicro + gradient	0.746	31.24	0.934
2	DeconvFormer + gradient	0.713	29.78	0.913
3	ResUNet + gradient	0.705	28.69	0.895
4	DiffDeconv + gradient	0.677	27.49	0.87
5	Restormer+ + gradient	0.671	26.98	0.858
6	Restormer + gradient	0.663	27.01	0.858
7	CARE + gradient	0.658	26.64	0.849
8	PnP-DnCNN + gradient	0.627	24.66	0.791
9	Wiener Filter + gradient	0.625	24.85	0.797
10	U-Net + gradient	0.623	25.0	0.802
11	TV-Deconvolution + gradient	0.623	24.32	0.78
12	Richardson-Lucy + gradient	0.581	22.7	0.719
13	PnP-FISTA + gradient	0.525	20.47	0.621

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
scattering_coeff	11.5	26.5	mm^-1
excitation_wavelength_shift	-1.4	4.6	nm
depth_dependent_psf	-0.28	0.92	-

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

C → D

C Convolution

D Detector

Mismatch Parameters

Symbol	Parameter	Description	Nominal	Perturbed
s_c	scattering_coeff	Scattering coeff (mm^-1)	15.0	20.0
e_w	excitation_wavelength_shift	Excitation wavelength shift (nm)	0.0	2.0
d_p	depth_dependent_psf	Depth-dependent PSF (-)	0.0	0.4

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.