Spinning Disk Confocal 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
🥇	DeconvFormer + gradient DeconvFormer + gradient Chen et al., CVPR 2024 Score 0.780 Correct & Reconstruct →	0.780	0.834 35.89 dB / 0.973	0.773 31.88 dB / 0.941	0.734 30.59 dB / 0.925	✓ Certified	Chen et al., CVPR 2024
🥈	Restormer+ + gradient Restormer+ + gradient Zamir et al., ICCV 2024 Score 0.774 Correct & Reconstruct →	0.774	0.817 35.02 dB / 0.968	0.778 32.51 dB / 0.948	0.727 29.39 dB / 0.907	✓ Certified	Zamir et al., ICCV 2024
🥉	ScoreMicro + gradient ScoreMicro + gradient Wei et al., ECCV 2025 Score 0.747 Correct & Reconstruct →	0.747	0.828 36.04 dB / 0.974	0.725 29.17 dB / 0.903	0.689 28.25 dB / 0.886	✓ Certified	Wei et al., ECCV 2025
4	ResUNet + gradient ResUNet + gradient DeCelle et al., Nat. Methods 2021 Score 0.732 Correct & Reconstruct →	0.732	0.793 33.12 dB / 0.954	0.722 29.33 dB / 0.906	0.680 26.85 dB / 0.854	✓ Certified	DeCelle et al., Nat. Methods 2021
5	Restormer + gradient Restormer + gradient Zamir et al., CVPR 2022 Score 0.729 Correct & Reconstruct →	0.729	0.815 34.04 dB / 0.961	0.727 28.91 dB / 0.899	0.644 25.15 dB / 0.807	✓ Certified	Zamir et al., CVPR 2022
6	DiffDeconv + gradient DiffDeconv + gradient Huang et al., NeurIPS 2024 Score 0.710 Correct & Reconstruct →	0.710	0.823 35.65 dB / 0.972	0.670 26.33 dB / 0.841	0.636 24.6 dB / 0.789	✓ Certified	Huang et al., NeurIPS 2024
7	U-Net + gradient U-Net + gradient Ronneberger et al., MICCAI 2015 Score 0.692 Correct & Reconstruct →	0.692	0.787 33.31 dB / 0.955	0.672 26.26 dB / 0.839	0.618 23.93 dB / 0.766	✓ Certified	Ronneberger et al., MICCAI 2015
8	CARE + gradient CARE + gradient Weigert et al., Nat. Methods 2018 Score 0.688 Correct & Reconstruct →	0.688	0.799 32.84 dB / 0.951	0.664 25.6 dB / 0.821	0.601 23.81 dB / 0.762	✓ Certified	Weigert et al., Nat. Methods 2018
9	TV-Deconvolution + gradient TV-Deconvolution + gradient Rudin et al., Phys. A 1992 Score 0.655 Correct & Reconstruct →	0.655	0.723 28.36 dB / 0.888	0.643 24.93 dB / 0.800	0.599 24.06 dB / 0.771	✓ Certified	Rudin et al., Phys. A 1992
10	PnP-FISTA + gradient PnP-FISTA + gradient Bai et al., 2020 Score 0.652 Correct & Reconstruct →	0.652	0.715 28.48 dB / 0.891	0.649 25.54 dB / 0.819	0.592 22.91 dB / 0.728	✓ Certified	Bai et al., 2020
11	PnP-DnCNN + gradient PnP-DnCNN + gradient Zhang et al., IEEE TIP 2017 Score 0.646 Correct & Reconstruct →	0.646	0.720 28.3 dB / 0.887	0.648 25.68 dB / 0.823	0.571 22.12 dB / 0.695	✓ Certified	Zhang et al., IEEE TIP 2017
12	Wiener Filter + gradient Wiener Filter + gradient Analytical baseline Score 0.637 Correct & Reconstruct →	0.637	0.672 26.15 dB / 0.836	0.638 24.63 dB / 0.790	0.600 23.38 dB / 0.746	✓ Certified	Analytical baseline
13	Richardson-Lucy + gradient Richardson-Lucy + gradient Richardson, JOSA 1972 / Lucy, AJ 1974 Score 0.620 Correct & Reconstruct →	0.620	0.675 25.82 dB / 0.827	0.623 24.42 dB / 0.783	0.562 22.6 dB / 0.715	✓ Certified	Richardson, JOSA 1972 / Lucy, AJ 1974

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	DeconvFormer + gradient	0.834	35.89	0.973
2	ScoreMicro + gradient	0.828	36.04	0.974
3	DiffDeconv + gradient	0.823	35.65	0.972
4	Restormer+ + gradient	0.817	35.02	0.968
5	Restormer + gradient	0.815	34.04	0.961
6	CARE + gradient	0.799	32.84	0.951
7	ResUNet + gradient	0.793	33.12	0.954
8	U-Net + gradient	0.787	33.31	0.955
9	TV-Deconvolution + gradient	0.723	28.36	0.888
10	PnP-DnCNN + gradient	0.720	28.3	0.887
11	PnP-FISTA + gradient	0.715	28.48	0.891
12	Richardson-Lucy + gradient	0.675	25.82	0.827
13	Wiener Filter + gradient	0.672	26.15	0.836

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
pinhole_crosstalk	-3.0	6.0	-
disk_rotation_wobble	-0.2	0.4	px
illumination_non_uniformity	-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	Restormer+ + gradient	0.778	32.51	0.948
2	DeconvFormer + gradient	0.773	31.88	0.941
3	Restormer + gradient	0.727	28.91	0.899
4	ScoreMicro + gradient	0.725	29.17	0.903
5	ResUNet + gradient	0.722	29.33	0.906
6	U-Net + gradient	0.672	26.26	0.839
7	DiffDeconv + gradient	0.670	26.33	0.841
8	CARE + gradient	0.664	25.6	0.821
9	PnP-FISTA + gradient	0.649	25.54	0.819
10	PnP-DnCNN + gradient	0.648	25.68	0.823
11	TV-Deconvolution + gradient	0.643	24.93	0.8
12	Wiener Filter + gradient	0.638	24.63	0.79
13	Richardson-Lucy + gradient	0.623	24.42	0.783

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
pinhole_crosstalk	-3.6	5.4	-
disk_rotation_wobble	-0.24	0.36	px
illumination_non_uniformity	-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	DeconvFormer + gradient	0.734	30.59	0.925
2	Restormer+ + gradient	0.727	29.39	0.907
3	ScoreMicro + gradient	0.689	28.25	0.886
4	ResUNet + gradient	0.680	26.85	0.854
5	Restormer + gradient	0.644	25.15	0.807
6	DiffDeconv + gradient	0.636	24.6	0.789
7	U-Net + gradient	0.618	23.93	0.766
8	CARE + gradient	0.601	23.81	0.762
9	Wiener Filter + gradient	0.600	23.38	0.746
10	TV-Deconvolution + gradient	0.599	24.06	0.771
11	PnP-FISTA + gradient	0.592	22.91	0.728
12	PnP-DnCNN + gradient	0.571	22.12	0.695
13	Richardson-Lucy + gradient	0.562	22.6	0.715

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
pinhole_crosstalk	-2.1	6.9	-
disk_rotation_wobble	-0.14	0.46	px
illumination_non_uniformity	-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

C → D

C Convolution

D Detector

Mismatch Parameters

Symbol	Parameter	Description	Perturbed
p_c	pinhole_crosstalk	Pinhole crosstalk (-)	3.0
d_r	disk_rotation_wobble	Disk rotation wobble (px)	0.2
i_n	illumination_non_uniformity	Illumination non-uniformity (-)	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.