AI Paper Atlas

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

Confidence-Weighted Local Expression Predictions for Occlusion Handling in Expression Recognition and Action Unit Detection

Arnaud Dapogny, Kévin Bailly, Séverine Dubuisson

IJCV 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Uncertainty-aware facial expression recognition · 79 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
make trees local (locality R=0.1) instead of global (ablation)RS-RF · CK+ (8 expressions)accuracy (%): 91.5 -> 93.4Improved
make trees local instead of global (ablation)RS-RF · BU4Daccuracy (%): 73.1 -> 74.3Improved
add autoencoder confidence weighting to local predictions (ablation)LS-RF · BU4Daccuracy (%): 74.3 -> 75.0Improved
add autoencoder confidence weighting to local predictions (ablation)LS-RF · SFEWaccuracy (%): 35.6 -> 37.1Improved
add autoencoder confidence weighting to local predictions (ablation)LS-RF · CK+ (8 expressions)accuracy (%): 93.4 -> 93.4No clear effect
add local trees and confidence weighting under mouth occlusion (ablation)global RS-RF · CK+ (8 expressions), mouth occludedaccuracy (%): 25.4 -> 67.1Improved
use WLS-RF instead of prior occlusion method under mouth occlusionWLS-RF · Mouth occlusionaccuracy (%): 30.3 -> 67.1Improved
merge categorical datasets (M2) instead of CK+ only for LEP training (ablation)AU detection · CK+ AUsaverage AUC (%): 93.4 -> 93.7Improved
merge categorical datasets (M2) instead of CK+ only for LEP training (ablation)AU detection · BP4D AUsaverage AUC (%): 70.8 -> 70.6No clear effect
merge datasets (M1) instead of CK+ only (ablation)AU detection · DISFA AUsaverage AUC (%): 72.9 -> 75.9Improved
train LEPs on SFEW only instead of CK+AU detection · CK+ AUsaverage AUC (%): 93.4 -> 88.9Hurt
use LEP features instead of SHTLAU detection · CK+ AUsaverage AUC (%): 91.7 -> 93.7Improved

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