A heterogeneity-aware framework with time-frequency attention and adaptive loss balancing reports state-of-the-art weighted and unweighted accuracy on DFEW and FERV39k.
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
-
Learning from Heterogeneity: Generalizing Dynamic Facial Expression Recognition via Distributionally Robust Optimization
A heterogeneity-aware framework with time-frequency attention and adaptive loss balancing reports state-of-the-art weighted and unweighted accuracy on DFEW and FERV39k.