CMNet is a new cross-modal network that uses face symmetry on full and half faces with refinement and alignment modules to improve facial expression recognition and outperforms SCN and LAENet-SA.
Facial expression recognition with inconsistently annotated datasets,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A cross-modal network for facial expression recognition
CMNet is a new cross-modal network that uses face symmetry on full and half faces with refinement and alignment modules to improve facial expression recognition and outperforms SCN and LAENet-SA.