AGCM classifies facial expressions and conversational engagement through learnable, spatially localized human-readable concepts, reporting accuracy at or above black-box baselines on RAF-DB, AffectNet, Aff-Wild2, and NOXI.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Interpretable Concept-based Deep Learning Framework for Multimodal Human Behavior Modeling
AGCM classifies facial expressions and conversational engagement through learnable, spatially localized human-readable concepts, reporting accuracy at or above black-box baselines on RAF-DB, AffectNet, Aff-Wild2, and NOXI.