Dev

High Dynamic Range (HDR) Imaging — 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
camera_response_function_error	-2.4 – 3.6	-
exposure_ratio_error	-2.4 – 3.6	-
ghost_artifact_(motion_between_exposures)	-1.2 – 1.8	px

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	DeblurGaussian + gradient	0.787	33.18	0.954	0.87	✓ Certified	Liang et al., CVPR 2024
2	PhotoFormer + gradient	0.779	32.25	0.945	0.89	✓ Certified	Zhang et al., ICCV 2024
3	HDRFormer + gradient	0.772	32.07	0.943	0.87	✓ Certified	Eilertsen et al., ICCV 2024
4	Uformer + gradient	0.730	30.06	0.918	0.81	✓ Certified	Wang et al., CVPR 2022
5	HDR-CNN + gradient	0.713	29.13	0.903	0.8	✓ Certified	Eilertsen et al., ACM TOG 2017
6	DiffusionPhoto + gradient	0.709	28.8	0.897	0.81	✓ Certified	Zhang et al., NeurIPS 2024
7	ScorePhoto + gradient	0.701	28.59	0.893	0.79	✓ Certified	Wei et al., ECCV 2025
8	U-Net + gradient	0.694	27.35	0.866	0.87	✓ Certified	Ronneberger et al., MICCAI 2015
9	PnP-ADMM + gradient	0.674	26.84	0.854	0.82	✓ Certified	Venkatakrishnan et al., 2013
10	LaplacianFormer + gradient	0.658	25.33	0.812	0.9	✓ Certified	Chen et al., CVPR 2022
11	Laplacian Pyramid + gradient	0.637	24.9	0.799	0.84	✓ Certified	Burt & Adelson, TPAMI 1983
12	PnP-FFDNet + gradient	0.636	24.79	0.795	0.85	✓ Certified	Zhang et al., 2017
13	Wiener-Deconv + gradient	0.634	25.06	0.804	0.81	✓ Certified	Analytical baseline
14	Lucy-Richardson + gradient	0.612	24.03	0.77	0.82	✓ Certified	Lucy, AJ 1974

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%