AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Metric Convolutions: A Unifying Theory to Adaptive Image Convolutions

Thomas Dagès, Michael Lindenbaum, Alfred M. Bruckstein

ICCV 2025 · Architectures and efficiency › CNN design · approach: Theory of convolutional inductive bias · 5 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
replace deformable convolution with metric UTB convolutionsingle-layer denoiser · single image denoising (sigma 0.3, k=5)normalised generalisation gap: 265 -> 1.3Improved
replace deformable convolution with metric UTB convolution at large kernelsingle-layer denoiser · single image denoising (sigma 0.3, k=121)test MSE: 6.18e-2 -> 7.82e-3Improved
increase metric asymmetry (eps_w 0.9 to 0.1) (ablation)metric UTB convolution · single image denoising (sigma 0.3, k=11)test MSE: 7.43e-3 -> 7.32e-3Improved
replace deformable with metric UTB convolution (fixed weights, scratch)ResNet18 · CIFAR-10top-1 accuracy: 33.21% -> 69.83%Improved
replace deformable with metric UTB convolution (fixed weights, scratch)ResNet18 · CIFAR-100top-1 accuracy: 10.47% -> 40.63%Improved
replace deformable with metric UTB convolution (learnable weights, scratch)ResNet18 · CIFAR-100top-1 accuracy: 54.51% -> 70.03%Improved
replace standard convolution with metric UTB (learnable weights, transfer)ResNet18 · CIFAR-10top-1 accuracy: 92.64% -> 93.07%Improved
replace standard convolution with metric UTB (learnable weights, transfer)ResNet18 · CIFAR-100top-1 accuracy: 70.52% -> 70.38%No clear effect
replace standard convolution with metric UTB (learnable weights, transfer)ResNet18 · Fashion-MNISTtop-1 accuracy: 93.30% -> 93.49%Improved
replace deformable with metric UTB on CIFAR-10 LKW-TLResNet18 · CIFAR-10top-1 accuracy: 93,10% -> 93.07%No clear effect
learn kernel weights instead of fixing them (from scratch) (ablation)metric UTB ResNet18 · CIFAR-10top-1 accuracy: 69.83% -> 91.83%Improved
use transfer learning instead of scratch (learnable weights) (ablation)metric UTB ResNet18 · CIFAR-100top-1 accuracy: 70.03% -> 70.38%No clear effect

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