AI Paper Atlas

Research map / Video / Action recognition

Order-Preserving Wasserstein Discriminant Analysis

Bing Su, Jiahuan Zhou, Ying Wu

ICCV 2019 · Video › Action recognition · approach: Contrastive video representation learning · 9 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 OWDA dimensionality reduction instead of LSDA before the SVM classifier (few dimensions)sequence DRS + SVM · MSR Action3Daccuracy and MAP: margin of about 10%Improved
add OWDA reduction to 30 dimensions before the SVM classifieroriginal 192-dim features + SVM · MSR Action3Daccuracy and MAP: better accuracy and comparable MAP using only 30 dimensionsImproved
add OWDA reduction to 25 dimensions before the SVM classifieroriginal features + SVM · MSR Daily Activity3DMAP: comparable MAP using only 25 dimensionsNo clear effect
apply OWDA dimensionality reduction before the SVM classifieroriginal features + SVM · MSR Daily Activity3DaccuracyHurt
use OWDA instead of other DRS methods with NN classifierNN classifier · MSR Daily Activity3DaccuracyHurt
use OWDA instead of other DRS methods with NN classifierNN classifier · MSR Daily Activity3DMAPImproved
use OWDA instead of other DRS methods with SVM classifierSVM classifier · ChaLearn Gestureaccuracy: margin of about 5%Improved
add OWDA to the original features with NN classifieroriginal features + NN · ChaLearn GestureMAP: margin of 20%Improved
use OWDA instead of LSDA with NN classifierNN classifier · ChaLearn GestureMAP: 5% higherImproved
use OWDA instead of LSDA with NN classifierNN classifier · ChaLearn GestureaccuracyNo clear effect
use OWDA+SVM instead of LSDA+SVMChaLearn pipeline · ChaLearn Gesture (validation)F-score: 0.767 -> 0.772Improved
apply OWDA (reduce to 80 dims) before Kernelized-COVKernelized-COV · MSR Daily Activity3Daccuracy: 96.3% -> 98.1%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