Hidden

High Dynamic Range (HDR) Imaging — 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.

Parameter	Spec Range	Unit
camera_response_function_error	-1.4 – 4.6	-
exposure_ratio_error	-1.4 – 4.6	-
ghost_artifact_(motion_between_exposures)	-0.7 – 2.3	px

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	DeblurGaussian + gradient	0.767	32.69	0.95	0.8	✓ Certified	Liang et al., CVPR 2024
2	HDRFormer + gradient	0.748	31.21	0.934	0.81	✓ Certified	Eilertsen et al., ICCV 2024
3	PhotoFormer + gradient	0.730	30.7	0.927	0.76	✓ Certified	Zhang et al., ICCV 2024
4	ScorePhoto + gradient	0.664	26.74	0.852	0.78	✓ Certified	Wei et al., ECCV 2025
5	HDR-CNN + gradient	0.663	26.22	0.838	0.83	✓ Certified	Eilertsen et al., ACM TOG 2017
6	U-Net + gradient	0.656	25.48	0.817	0.87	✓ Certified	Ronneberger et al., MICCAI 2015
7	DiffusionPhoto + gradient	0.654	25.4	0.815	0.87	✓ Certified	Zhang et al., NeurIPS 2024
8	Uformer + gradient	0.636	24.87	0.798	0.84	✓ Certified	Wang et al., CVPR 2022
9	LaplacianFormer + gradient	0.604	24.01	0.769	0.78	✓ Certified	Chen et al., CVPR 2022
10	PnP-ADMM + gradient	0.604	23.63	0.755	0.83	✓ Certified	Venkatakrishnan et al., 2013
11	Wiener-Deconv + gradient	0.601	23.33	0.744	0.85	✓ Certified	Analytical baseline
12	Laplacian Pyramid + gradient	0.595	23.03	0.732	0.86	✓ Certified	Burt & Adelson, TPAMI 1983
13	PnP-FFDNet + gradient	0.572	22.21	0.699	0.85	✓ Certified	Zhang et al., 2017
14	Lucy-Richardson + gradient	0.545	22.05	0.692	0.74	✓ Certified	Lucy, AJ 1974

Scenes: 5

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

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