AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Integrating Multiple Receptive Fields Through Grouped Active Convolution

Yunho Jeon, Junmo Kim

TPAMI 2020 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 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 3x3 conv with ACUResNet-29 bottleneck · CIFAR-10test error (%): 6.16 -> 5.56Improved
enlarge receptive field via dilation 2 in last stage (ablation)ResNet-29 baseline · CIFAR-10test error (%): 6.16 -> 6.28Hurt
fix ACU shape to learned shape and retrain weights (ablation)ACU network · CIFAR-10test error (%): 5.56 -> 5.44Improved
replace grouped conv with grouped ACU16x4d ResNet · CIFAR-10test error (%): 5.38 -> 4.93Improved
increase cardinality (4d to 1d) with naive 3x3 conv (ablation)grouped conv network · CIFAR-10test error (%): 5.38 -> 5.75Hurt
increase cardinality (4d to 1d) with ACU (ablation)grouped ACU network · CIFAR-10test error (%): 4.93 -> 4.9No clear effect
add warming-up for shared-position ACU (ablation)ResNet/ACU · CIFAR-10test error (%): 5.64 -> 5.56Improved
add warming-up for depthwise ACU (ablation)x1d/ACU · CIFAR-10test error (%): 4.9 -> 5.04Hurt
use multiple position sets instead of one shared (ablation)depthwise ACU x1d · CIFAR-10test error (%): 5.2 -> 4.9Improved
replace Inception blocks with 1x1 conv plus depthwise ACU blockInception+BN · ImageNettop-1 accuracy (%): 72.0 -> 72.1Improved
replace last depthwise conv per block with depthwise ACUXception · ImageNettop-1 accuracy (%): 75.6 -> 75.9Improved
replace depthwise conv with depthwise ACUMobileNet v2 x1.0 · ImageNettop-1 accuracy (%): 71.3 -> 72.1Improved

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