Hidden

X-ray Radiography — Hidden Tier

(3 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	CTFormer + gradient	0.770	33.68	0.958	0.75	✓ Certified	Li et al., ICCV 2024
2	CT-ViT + gradient	0.767	32.42	0.947	0.82	✓ Certified	Guo et al., NeurIPS 2024
3	DOLCE + gradient	0.711	28.72	0.895	0.83	✓ Certified	Liu et al., ICCV 2023
4	DiffusionCT + gradient	0.702	28.19	0.885	0.83	✓ Certified	Kazemi et al., ECCV 2024
5	Score-CT + gradient	0.691	27.94	0.88	0.8	✓ Certified	Song et al., NeurIPS 2024
6	DuDoTrans + gradient	0.686	27.64	0.873	0.8	✓ Certified	Wang et al., MLMIR 2022
7	PnP-ADMM + gradient	0.673	26.29	0.84	0.87	✓ Certified	Venkatakrishnan et al., IEEE GlobalSIP 2013
8	RED-CNN + gradient	0.666	26.43	0.844	0.82	✓ Certified	Chen et al., IEEE TMI 2017
9	Learned Primal-Dual + gradient	0.639	25.74	0.825	0.76	✓ Certified	Adler & Oktem, IEEE TMI 2018
10	TV-ADMM + gradient	0.633	25.2	0.808	0.79	✓ Certified	Sidky et al., Phys. Med. Biol. 2008
11	PnP-DnCNN + gradient	0.630	24.96	0.801	0.8	✓ Certified	Zhang et al., IEEE TIP 2017
12	FBP + gradient	0.573	22.64	0.717	0.8	✓ Certified	Kak & Slaney, IEEE Press 1988
13	FBPConvNet + gradient	0.560	21.94	0.687	0.83	✓ Certified	Jin et al., IEEE TIP 2017

Scenes: 3

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

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