Dev

X-ray Radiography — Dev Tier

(3 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.

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	CTFormer + gradient	0.800	35.44	0.97	0.79	✓ Certified	Li et al., ICCV 2024
2	CT-ViT + gradient	0.789	34.17	0.962	0.81	✓ Certified	Guo et al., NeurIPS 2024
3	DOLCE + gradient	0.755	30.79	0.928	0.88	✓ Certified	Liu et al., ICCV 2023
4	DiffusionCT + gradient	0.745	30.38	0.922	0.86	✓ Certified	Kazemi et al., ECCV 2024
5	Score-CT + gradient	0.742	30.56	0.925	0.83	✓ Certified	Song et al., NeurIPS 2024
6	DuDoTrans + gradient	0.723	28.83	0.897	0.88	✓ Certified	Wang et al., MLMIR 2022
7	Learned Primal-Dual + gradient	0.694	27.46	0.869	0.86	✓ Certified	Adler & Oktem, IEEE TMI 2018
8	PnP-DnCNN + gradient	0.692	27.78	0.876	0.82	✓ Certified	Zhang et al., IEEE TIP 2017
9	PnP-ADMM + gradient	0.690	27.98	0.88	0.79	✓ Certified	Venkatakrishnan et al., IEEE GlobalSIP 2013
10	RED-CNN + gradient	0.689	27.0	0.858	0.88	✓ Certified	Chen et al., IEEE TMI 2017
11	TV-ADMM + gradient	0.669	25.9	0.829	0.89	✓ Certified	Sidky et al., Phys. Med. Biol. 2008
12	FBPConvNet + gradient	0.663	25.62	0.821	0.89	✓ Certified	Jin et al., IEEE TIP 2017
13	FBP + gradient	0.621	24.34	0.78	0.83	✓ Certified	Kak & Slaney, IEEE Press 1988

y H_ideal spec_ranges

Format: HDF5

Scenes: 3

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

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