AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Second-order Democratic Aggregation

Tsung-Yu Lin, Subhransu Maji, Piotr Koniusz

ECCV 2018 · Architectures and efficiency › CNN design · approach: Higher-order pooling · 33 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
use fully democratic pooling (gamma=0) instead of sum pooling (ablation)second-order features, VGG-16 · Birds, Aircrafts, DTD, MIT indooraccuracy (%): 0.7-1%Improved
use fully democratic pooling (gamma=0) instead of sum pooling (ablation)second-order features, VGG-16 · Stanford Cars, FMDaccuracy (%): 90.6 -> 89.7; 84.6 -> 82.8Hurt
tune gamma between 0 and 1 instead of sum pooling (gamma=1) (ablation)gamma-democratic pooling · Caltech UCSD Birdsaccuracy (%): 84.0 -> 84.9Improved
tune gamma between 0 and 1 instead of sum pooling (gamma=1) (ablation)gamma-democratic pooling · FGVC Aircraftsaccuracy (%): 85.7 -> 86.7Improved
tune gamma instead of sum pooling (ablation)gamma-democratic pooling · DTDaccuracy (%): 71.2 -> 72.3Improved
tune gamma instead of sum pooling (ablation)gamma-democratic pooling · FMDaccuracy (%): 84.6 -> 84.8No clear effect
tune gamma instead of sum pooling (ablation)gamma-democratic pooling · Stanford Carsaccuracy (%): 90.6 -> 90.8Improved
tune gamma between 0 and 1 instead of democratic pooling (gamma=0) (ablation)gamma-democratic pooling · FMDaccuracy (%): 82.8 -> 84.8Improved
use matrix power normalization instead of gamma-democratic pooling (ablation)second-order features, VGG-16 · six datasetsaccuracy (%): 0.2-1%Improved
replace sum pooling with gamma-democratic aggregation on tensor-sketched ResNet101 features (ablation)ResNet101 + TS · MIT indooraccuracy (%): 82.8 -> 84.3Improved
use tensor-sketched gamma-democratic ResNet101 instead of matrix power normalization with VGG-16gamma-democratic · MIT indooraccuracy (%): +3.4%Improved

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