AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Global Feature Guided Local Pooling

Takumi Kobayashi

ICCV 2019 · Architectures and efficiency › CNN design · approach: Higher-order pooling · 27 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
estimate lambda from global features instead of fixed constant (ablation)parameterized pooling · ImageNet val (VGG-13)top-1 error (%): 29.82 -> 28.57Improved
estimate lambda from global features instead of training it as a parameter (ablation)parameterized pooling · ImageNet val (VGG-13)top-1 error (%): 29.56 -> 28.57Improved
use global average instead of local 2x2 average for lambda estimation (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 29.02 -> 28.57Improved
add estimated position priors rho (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.57 -> 28.62No clear effect
use global avg statistic instead of max (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.93 -> 28.57Improved
add std to global avg statistic (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.57 -> 28.56No clear effect
use MLP-1/2 instead of single layer perceptron (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.72 -> 28.57Improved
double hidden nodes (MLP-1 vs MLP-1/2) (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.57 -> 28.36Improved
use sigmoid instead of tanh activation for lambda (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 28.96 -> 28.57Improved
GFGP vs NiN module with same extra parameters (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 30.61 -> 28.57Improved
GFGP vs SE module with same extra parameters (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 29.24 -> 28.57Improved
replace all five pooling layers instead of only the first (ablation)GFGP · ImageNet val (VGG-13)top-1 error (%): 30.01 -> 28.57Improved

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