AI Paper Atlas

Research map / Image generation / Image restoration

Fourier Series Expansion Based Filter Parametrization for Equivariant Convolutions

Qi Xie, Qian Zhao, Zongben Xu, Deyu Meng

TPAMI 2022 · Image generation › Image restoration · approach: Real-world restoration with unaligned data · 35 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use proposed Fourier-based bases instead of traditional 2D Fourier bases for filter parametrizationF-Conv · Filter parametrization 5x5 (0 and 45 deg simultaneously)RMSE: 1.8e-01 -> 6.6e-02Improved
use proposed bases instead of harmonic basesF-Conv · Filter parametrization 11x11 (0 and 45 deg simultaneously)RMSE: 2.7e-01 -> 2.0e-02Improved
use proposed bases instead of 2D Fourier bases on random initial filtersF-Conv · Random filter parametrization 11x11RMSE: 9.5e-13No clear effect
use F-Conv instead of E2-CNN for equivariance error (trained)F-Conv · Equivariance error DIV2K (trained)RMSE: 0.18 -> 0.19No clear effect
use F-Conv instead of E2-CNN for equivariance error (random init)F-Conv · Equivariance error DIV2K (random init)RMSE: 0.30 -> 0.42Hurt
use F-Conv instead of plain CNN for equivariance error (random init)F-Conv · Equivariance error DIV2K (random init)RMSE: 1.43 -> 0.42Improved
use all 25 proposed bases instead of first 11 principal components (ablation)F-Conv · MNIST-rot-12k (simple 6-layer network)test error (%): 1.23 -> 1.13Improved
use F-Conv-reduce instead of E2-CNN at equal parametersF-Conv-reduce · MNIST-rot-12k (simple 6-layer network)test error (%): 1.23 -> 1.23No clear effect
use F-Conv instead of E2-CNN in larger MNIST-rot-12k networkF-Conv · MNIST-rot-12k (leaderboard comparison)test error (%): 0.682 -> 0.671Improved
use F-Conv instead of CNN for EDSR x4 SR without augmentationF-Conv · SR x4 (4 datasets avg, no aug)PSNR (dB): 26.95 -> 27.15Improved
use E2-CNN instead of CNN in EDSR x2 SRE2-CNN · SR x2 (4 datasets avg, no aug)PSNR (dB): 32.39 -> 32.23Hurt
use PDO-eConv instead of CNN in EDSR x2 SRPDO-eConv · SR x2 (4 datasets avg, no aug)PSNR (dB): 32.39 -> 31.62Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy