UF-AMA: A unified framework for cross-domain emotion recognition via adaptive multimodal alignment
Researchers have developed UF-AMA, a novel framework designed to improve emotion recognition across different datasets and sessions using physiological signals. This unified approach integrates EEG and eye-tracking data through Transformer encoders and cross-attention modules. It incorporates a confidence-aware mechanism to assess predictive reliability and adaptively aligns multimodal distributions, demonstrating state-of-the-art performance on benchmark datasets. AI
IMPACT This framework could lead to more robust and generalizable AI systems for understanding human emotions from physiological data.