AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Adaptive Aggregation of Monte Carlo Augmented Decomposed Filters for Efficient Group-Equivariant Convolutional Neural Network

Wenkai Zhao, Barbara Daria Wichtmann, Steffen Albert, Angelika Maurer, Frank Gerrit Zöllner, Jürgen Hesser

TPAMI 2026 · Architectures and efficiency › CNN design · approach: Group convolution theory · 2 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
add shear augmentation range of 0.12pi to FB basis (ablation)WMCG ResNet18 5x5 · ImageNet40error (%): 19.60 -> 18.80Improved
add shear augmentation range of 0.25pi to FB basis (ablation)WMCG ResNeXt50 5x5 · ImageNet40error (%): 27.60 -> 26.95Improved
increase shear range from 0.12pi to 0.25pi on MH basis (ablation)WMCG ResNet18 3x3 MH · ImageNet40error (%): 24.55 -> 26.85Hurt
use high-order (8th) Fourier-Bessel basis instead of 1st (ablation)WMCG ResNet18 single FB basis · ImageNet40error (%): 28.65 -> 55.45Hurt
use MC affine sampling instead of MC scale-only sampling (16 samples, width 1/16) (ablation)MCG-CNN ResNet18 · ImageNet40error (%): 36.75 -> 32.65Improved
use MC instead of equally spaced sampling (scale, 16 samples, width 1/4) (ablation)MCG-CNN ResNet18 · ImageNet40error (%): 19.00 -> 19.20No clear effect
replace conv layers with WMCG 5x5 layersResNet50 (Pixmix) · ImageNet1k-CmCE (%): 54.23 -> 53.04Improved
replace 7x7 conv layers with WMCGConvNeXt-S · ImageNet40top-1 accuracy (%): 84.75 -> 86.65Improved
replace 7x7 conv layers with dynamic convolutionConvNeXt-S · ImageNet1ktop-1 accuracy (%): 83.14 -> 82.57Hurt
replace conv layers with WMCG 5x5 layersDudeNeXt-B · color denoising averagePSNR (dB): 31.35 -> 31.45Improved
enlarge WMCG kernel from 5x5 to 7x7 (ablation)DnNeXt-B · grayscale denoising averagePSNR (dB): 29.26 -> 29.20Mixed
replace conv layers with WMCGResNeXt41 · STL10 OODaccuracy (%): 81.13 -> 83.71Improved

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