AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks

Jie Hu, Li Shen, Samuel Albanie, Gang Sun, Andrea Vedaldi

NeurIPS 2018 · Architectures and efficiency › CNN design · approach: Filter design and trained-filter analysis · 499 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 parameter-free gather-excite context modulationResNet-50 · ImageNet-1K valtop-1 error (%): 23.30 -> 22.14Improved
Add parameterized gather and excite operatorsResNet-50 · ImageNet-1K valtop-1 error (%): 22.00 -> 21.88Improved
Add GE-θ operators in stage 2 only (ablation)ResNet-50 · ImageNet-1K valtop-1 error (%): 23.30 -> 23.29No clear effect
Add GE-θ operators in stage 3 only (ablation)ResNet-50 · ImageNet-1K valtop-1 error (%): 23.30 -> 22.70Improved
Add GE-θ operators in stage 4 only (ablation)ResNet-50 · ImageNet-1K valtop-1 error (%): 23.30 -> 22.50Improved
Add GE operators in all stages rather than a single stage (ablation)ResNet-50 · ImageNet-1K valtop-1 error (%): 22.50 -> 22.00Improved
Increase gather extent ratio (ablation)GE-θ⁻ on ResNet-50 · ImageNet-1K valtop-1 error (%)Improved
Use global max pooling instead of average pooling (ablation)GE-θ⁻ on ResNet-50 · ImageNet-1K valtop-1 error (%): 22.14 -> 23.65Hurt
Use average pooling instead of max pooling (ablation)GE-θ⁻ on ResNet-50 · ImageNet-1K valtop-1 error (%): 23.10 -> 22.87; 23.65 -> 22.14Improved
Add parameterized gather-excite modulesShuffleNet · ImageNet-1K valtop-1 error (%): 32.60 -> 31.80Improved
Add GE-θ+ modules (ablation)ShuffleNet · ImageNet-1K valtop-1 error (%): 31.80 -> 30.12Improved
Add GE-θ+ modulesFaster R-CNN with ResNet-50 · MS COCO 2014 valmAP: 27.3 -> 28.6Improved

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