Lattice Light-Sheet 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 0.776 Correct & Reconstruct →	0.776	0.811 34.49 dB / 0.964	0.775 33.46 dB / 0.957	0.743 30.61 dB / 0.926	✓ Certified	Chen et al., CVPR 2024
🥈	Restormer + gradient Restormer + gradient Zamir et al., CVPR 2022 Score 0.753 Correct & Reconstruct →	0.753	0.796 33.82 dB / 0.959	0.759 30.92 dB / 0.930	0.704 28.85 dB / 0.898	✓ Certified	Zamir et al., CVPR 2022
🥉	ScoreMicro + gradient ScoreMicro + gradient Wei et al., ECCV 2025 Score 0.743 Correct & Reconstruct →	0.743	0.830 36.52 dB / 0.976	0.728 29.24 dB / 0.904	0.672 27.48 dB / 0.869	✓ Certified	Wei et al., ECCV 2025
4	Restormer+ + gradient Restormer+ + gradient Zamir et al., ICCV 2024 Score 0.741 Correct & Reconstruct →	0.741	0.839 36.48 dB / 0.976	0.743 30.25 dB / 0.921	0.642 26.06 dB / 0.834	✓ Certified	Zamir et al., ICCV 2024
5	DiffDeconv + gradient DiffDeconv + gradient Huang et al., NeurIPS 2024 Score 0.723 Correct & Reconstruct →	0.723	0.845 36.95 dB / 0.978	0.699 28.16 dB / 0.884	0.626 24.27 dB / 0.778	✓ Certified	Huang et al., NeurIPS 2024
6	U-Net + gradient U-Net + gradient Ronneberger et al., MICCAI 2015 Score 0.693 Correct & Reconstruct →	0.693	0.808 33.83 dB / 0.960	0.667 26.6 dB / 0.848	0.604 23.64 dB / 0.755	✓ Certified	Ronneberger et al., MICCAI 2015
7	ResUNet + gradient ResUNet + gradient DeCelle et al., Nat. Methods 2021 Score 0.693 Correct & Reconstruct →	0.693	0.794 33.56 dB / 0.957	0.687 27.4 dB / 0.868	0.598 23.27 dB / 0.742	✓ Certified	DeCelle et al., Nat. Methods 2021
8	TV-Deconvolution + gradient TV-Deconvolution + gradient Rudin et al., Phys. A 1992 Score 0.671 Correct & Reconstruct →	0.671	0.690 26.85 dB / 0.854	0.682 26.94 dB / 0.857	0.640 25.48 dB / 0.817	✓ Certified	Rudin et al., Phys. A 1992
9	Wiener Filter + gradient Wiener Filter + gradient Analytical baseline Score 0.665 Correct & Reconstruct →	0.665	0.699 26.87 dB / 0.855	0.654 26.04 dB / 0.833	0.643 25.56 dB / 0.819	✓ Certified	Analytical baseline
10	CARE + gradient CARE + gradient Weigert et al., Nat. Methods 2018 Score 0.662 Correct & Reconstruct →	0.662	0.797 32.74 dB / 0.950	0.642 24.71 dB / 0.793	0.547 21.39 dB / 0.663	✓ Certified	Weigert et al., Nat. Methods 2018
11	PnP-DnCNN + gradient PnP-DnCNN + gradient Zhang et al., IEEE TIP 2017 Score 0.659 Correct & Reconstruct →	0.659	0.754 30.2 dB / 0.920	0.629 25.01 dB / 0.803	0.593 22.95 dB / 0.729	✓ Certified	Zhang et al., IEEE TIP 2017
12	PnP-FISTA + gradient PnP-FISTA + gradient Bai et al., 2020 Score 0.657 Correct & Reconstruct →	0.657	0.706 27.53 dB / 0.871	0.642 24.97 dB / 0.801	0.623 24.05 dB / 0.770	✓ Certified	Bai et al., 2020
13	Richardson-Lucy + gradient Richardson-Lucy + gradient Richardson, JOSA 1972 / Lucy, AJ 1974 Score 0.624 Correct & Reconstruct →	0.624	0.636 24.35 dB / 0.781	0.641 24.99 dB / 0.802	0.594 23.6 dB / 0.754	✓ 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	DiffDeconv + gradient	0.845	36.95	0.978
2	Restormer+ + gradient	0.839	36.48	0.976
3	ScoreMicro + gradient	0.830	36.52	0.976
4	DeconvFormer + gradient	0.811	34.49	0.964
5	U-Net + gradient	0.808	33.83	0.96
6	CARE + gradient	0.797	32.74	0.95
7	Restormer + gradient	0.796	33.82	0.959
8	ResUNet + gradient	0.794	33.56	0.957
9	PnP-DnCNN + gradient	0.754	30.2	0.92
10	PnP-FISTA + gradient	0.706	27.53	0.871
11	Wiener Filter + gradient	0.699	26.87	0.855
12	TV-Deconvolution + gradient	0.690	26.85	0.854
13	Richardson-Lucy + gradient	0.636	24.35	0.781

Spec Ranges (4 parameters)

Parameter	Min	Max	Unit
lattice_period_error	-1.0	2.0	relative
dithering_range	-0.15	0.15	-
sheet_na_error	-0.01	0.02	-
excitation_psf_sidelobe	-2.0	4.0	relative

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	0.775	33.46	0.957
2	Restormer + gradient	0.759	30.92	0.93
3	Restormer+ + gradient	0.743	30.25	0.921
4	ScoreMicro + gradient	0.728	29.24	0.904
5	DiffDeconv + gradient	0.699	28.16	0.884
6	ResUNet + gradient	0.687	27.4	0.868
7	TV-Deconvolution + gradient	0.682	26.94	0.857
8	U-Net + gradient	0.667	26.6	0.848
9	Wiener Filter + gradient	0.654	26.04	0.833
10	CARE + gradient	0.642	24.71	0.793
11	PnP-FISTA + gradient	0.642	24.97	0.801
12	Richardson-Lucy + gradient	0.641	24.99	0.802
13	PnP-DnCNN + gradient	0.629	25.01	0.803

Spec Ranges (4 parameters)

Parameter	Min	Max	Unit
lattice_period_error	-1.2	1.8	relative
dithering_range	-0.15	0.15	-
sheet_na_error	-0.012	0.018	-
excitation_psf_sidelobe	-2.4	3.6	relative

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	0.743	30.61	0.926
2	Restormer + gradient	0.704	28.85	0.898
3	ScoreMicro + gradient	0.672	27.48	0.869
4	Wiener Filter + gradient	0.643	25.56	0.819
5	Restormer+ + gradient	0.642	26.06	0.834
6	TV-Deconvolution + gradient	0.640	25.48	0.817
7	DiffDeconv + gradient	0.626	24.27	0.778
8	PnP-FISTA + gradient	0.623	24.05	0.77
9	U-Net + gradient	0.604	23.64	0.755
10	ResUNet + gradient	0.598	23.27	0.742
11	Richardson-Lucy + gradient	0.594	23.6	0.754
12	PnP-DnCNN + gradient	0.593	22.95	0.729
13	CARE + gradient	0.547	21.39	0.663

Spec Ranges (4 parameters)

Parameter	Min	Max	Unit
lattice_period_error	-0.7	2.3	relative
dithering_range	-0.15	0.15	-
sheet_na_error	-0.007	0.023	-
excitation_psf_sidelobe	-1.4	4.6	relative

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
l_p	lattice_period_error	Lattice period error (relative)	1.0
d_r	dithering_range	Dithering range (-)	0.0
s_n	sheet_na_error	Sheet NA error (-)	0.01
e_p	excitation_psf_sidelobe	Excitation PSF sidelobe (relative)	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.