Dev

EELS — Dev Tier

(3 scenes)

Blind evaluation tier — no ground truth available.

What you get

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

How to use

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

What to submit

Reconstructed signals and corrected spec. Scored server-side.

Parameter Specifications

🔒

True spec hidden — estimate parameters from spec ranges below.

Parameter Spec Range Unit
energy_dispersion -0.0024 – 0.0036 eV/channel
zero_loss_shift -0.36 – 0.54 eV
aberration -2.4 – 3.6 %

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SwinEELS + gradient 0.767 31.84 0.941 0.86 ✓ Certified Wang et al., npj Comput. Mater. 2023
2 TransEELS + gradient 0.752 31.47 0.937 0.81 ✓ Certified Li et al., Ultramicroscopy 2022
3 PhysEELS + gradient 0.739 30.78 0.928 0.8 ✓ Certified Chen et al., Microsc. Microanal. 2024
4 DiffEELS + gradient 0.739 30.29 0.921 0.84 ✓ Certified Gao et al., NeurIPS 2024
5 N2V-EELS + gradient 0.612 23.43 0.748 0.89 ✓ Certified Krull et al., NeurIPS 2019
6 DnCNN-EELS + gradient 0.517 20.17 0.607 0.86 ✓ Certified Kovarik et al., npj Comput. Mater. 2016
7 PowerLaw-EELS + gradient 0.485 19.02 0.551 0.87 ✓ Certified Egerton, EELS in the EM, Springer 2011
8 MLS-EELS + gradient 0.429 17.62 0.481 0.8 ✓ Certified Verbeeck & Van Aert, Ultramicroscopy 2004
9 ICA-EELS + gradient 0.412 16.92 0.446 0.82 ✓ Certified Bosman et al., Ultramicroscopy 2006

Visible Data Fields

y H_ideal spec_ranges

Dataset

Format: HDF5
Scenes: 3

Scoring Formula

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

PSNR: 40% SSIM: 40% Consistency: 20%
Back to EELS