Hidden

Widefield Low-Dose — Hidden Tier

(5 scenes)

Fully blind server-side evaluation — no data download.

No data downloadable. Algorithm runs server-side on hidden measurements.

Package algorithm as Docker container / Python script. Submit via link.

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

🔒

True spec hidden — blind evaluation, only ranges available.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	DeconvFormer + gradient	0.757	31.19	0.933	0.86	✓ Certified	Chen et al., CVPR 2024
2	ScoreMicro + gradient	0.721	29.62	0.911	0.8	✓ Certified	Wei et al., ECCV 2025
3	DiffDeconv + gradient	0.715	28.82	0.897	0.84	✓ Certified	Huang et al., NeurIPS 2024
4	Restormer+ + gradient	0.691	27.17	0.862	0.87	✓ Certified	Zamir et al., ICCV 2024
5	ResUNet + gradient	0.689	27.84	0.877	0.8	✓ Certified	DeCelle et al., Nat. Methods 2021
6	TV-Deconvolution + gradient	0.664	26.73	0.851	0.78	✓ Certified	Rudin et al., Phys. A 1992
7	U-Net + gradient	0.656	25.88	0.829	0.83	✓ Certified	Ronneberger et al., MICCAI 2015
8	Restormer + gradient	0.643	25.83	0.827	0.77	✓ Certified	Zamir et al., CVPR 2022
9	Wiener Filter + gradient	0.615	24.39	0.782	0.79	✓ Certified	Analytical baseline
10	PnP-DnCNN + gradient	0.608	24.17	0.775	0.78	✓ Certified	Zhang et al., IEEE TIP 2017
11	CARE + gradient	0.606	23.51	0.751	0.85	✓ Certified	Weigert et al., Nat. Methods 2018
12	PnP-FISTA + gradient	0.606	23.32	0.743	0.88	✓ Certified	Bai et al., 2020
13	Richardson-Lucy + gradient	0.541	21.26	0.657	0.83	✓ Certified	Richardson, JOSA 1972 / Lucy, AJ 1974

Scenes: 5

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%