AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Distance Based Image Classification: A solution to generative classification’s conundrum?

Wen-Yan Lin, Siying Liu, Bing Tian Dai, Hongdong Li

IJCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 4 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 Soft-Max/discriminative layer with strictly incremental distance classifierResNet-50 features · STL-10classification accuracy: 0.967 -> 0.970Improved
replace kernel SVM with distance classifierResNet-50 features / pixels · MNISTclassification accuracy: 0.966 -> 0.955Hurt
replace kernel SVM with distance classifierpixels · Fashion-MNISTclassification accuracy: 0.884 -> 0.825Hurt
use distance classifier instead of Shell-R strict incremental learnerstrict incremental learners · STL-10classification accuracy: 0.930 -> 0.970Improved
use distance classification vs best replay-based baselineincremental learners · SUN-397 (top-5)final accuracy: 72.6 -> 87.5Improved
use distance classification vs replay-based baselinesincremental learners · ImageNet-1000 (top-5)final accuracy: 73.2 -> 88.0Improved
evaluate with corrected ImageNet labelsDistance classification · ImageNet-1000 (top-5)final accuracy: 88.0 -> 91.7Improved
use distance classification vs iTAMLincremental learners · Food-101 (top-5)final accuracy: 84.7 -> 81.1Hurt
use distance classification vs REMINDincremental learners · SUN-397 incremental updatetraining time for last increment (s): 2452 -> 453Improved
use distance classification vs REMINDincremental learners · Flower-102 incremental updatetraining time for last increment (s): 19 -> 38Hurt
add normalization to Euclidean nearest neighbor (ablation)Euc. NN · STL-10validation AUPRC: 0.764 -> 0.922Improved
use distance classifier instead of raw Euclidean NN (ablation)strict incremental · STL-10validation AUPRC: 0.764 -> 0.984Improved

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