Dev

Image Scanning Microscopy (ISM) — Dev Tier

(5 scenes)

Blind evaluation tier — no ground truth available.

Measurements (y), ideal forward operator (H), and spec ranges only.

Apply your pipeline from the Public tier. Use consistency as self-check.

Reconstructed signals and corrected spec. Scored server-side.

🔒

True spec hidden — estimate parameters from spec ranges below.

Parameter	Spec Range	Unit
detector_element_offset	-0.24 – 0.36	px
magnification_error	-1.2 – 1.8	relative

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	Restormer+ + gradient	0.767	31.89	0.941	0.86	✓ Certified	Zamir et al., ICCV 2024
2	DeconvFormer + gradient	0.759	31.99	0.943	0.81	✓ Certified	Chen et al., CVPR 2024
3	Restormer + gradient	0.748	30.85	0.929	0.84	✓ Certified	Zamir et al., CVPR 2022
4	ResUNet + gradient	0.719	28.34	0.888	0.9	✓ Certified	DeCelle et al., Nat. Methods 2021
5	ScoreMicro + gradient	0.713	29.16	0.903	0.8	✓ Certified	Wei et al., ECCV 2025
6	DiffDeconv + gradient	0.707	29.05	0.901	0.78	✓ Certified	Huang et al., NeurIPS 2024
7	CARE + gradient	0.698	28.08	0.882	0.82	✓ Certified	Weigert et al., Nat. Methods 2018
8	TV-Deconvolution + gradient	0.669	26.2	0.838	0.86	✓ Certified	Rudin et al., Phys. A 1992
9	PnP-DnCNN + gradient	0.664	26.05	0.833	0.85	✓ Certified	Zhang et al., IEEE TIP 2017
10	U-Net + gradient	0.662	26.16	0.836	0.83	✓ Certified	Ronneberger et al., MICCAI 2015
11	Wiener Filter + gradient	0.636	24.68	0.792	0.86	✓ Certified	Analytical baseline
12	Richardson-Lucy + gradient	0.626	24.51	0.786	0.83	✓ Certified	Richardson, JOSA 1972 / Lucy, AJ 1974
13	PnP-FISTA + gradient	0.622	24.53	0.787	0.81	✓ Certified	Bai et al., 2020

y H_ideal spec_ranges

Format: HDF5

Scenes: 5

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

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