SSM framework achieves simultaneous state-of-the-art results on AU detection and FE recognition by using textual semantic prototypes and dynamic prior mapping for bidirectional transfer across heterogeneous data.
Facial action unit detection with transform- ers,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
MEDN decouples explicit motion and implicit emotion features with a dual-branch design, AU restriction, orthogonal loss, SEVit, and adaptive fusion to improve micro-expression recognition on benchmarks.
citing papers explorer
-
Bidirectional Learning of Facial Action Units and Expressions via Structured Semantic Mapping across Heterogeneous Datasets
SSM framework achieves simultaneous state-of-the-art results on AU detection and FE recognition by using textual semantic prototypes and dynamic prior mapping for bidirectional transfer across heterogeneous data.
-
MEDN: Motion-Emotion Feature Decoupling Network for Micro-Expression Recognition
MEDN decouples explicit motion and implicit emotion features with a dual-branch design, AU restriction, orthogonal loss, SEVit, and adaptive fusion to improve micro-expression recognition on benchmarks.