A two-stream network that hides faces and attends to scene context, trained on a new CAER video dataset, improves emotion recognition accuracy over face-only models.
Learning active facial patches for expression analysis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Context-Aware Emotion Recognition Networks
A two-stream network that hides faces and attends to scene context, trained on a new CAER video dataset, improves emotion recognition accuracy over face-only models.