Our laboratory published an article in the international journal Pattern Recognition Letters (Scopus, Q1):

Ryumina E., Ryumin D., Axyonov A., Ivanko D., Karpov A. Multi-corpus emotion recognition method based on cross-modal gated attention fusion // Pattern Recognition Letters, 2025, vol. 190, pp. 192–200. (WOS IF=3.3 Q2, Scopus SJR=1.00 Q1 CV & PR; Q2 AI)

This paper proposes the first multi-corpus multimodal emotion recognition method with high generalizability evaluated under a leave-one-corpus-out protocol. The method uses three fine-tuned encoders per modality (audio, video, and text) and a decoder employing context-independent gated attention to combine features from all three modalities. The research is conducted on four benchmark corpora: MOSEI, MELD, IEMOCAP, and AFEW. The proposed method achieves state-of-the-art results on these corpora and establishes the first baseline for multi-corpus studies. We show that due to MELD’s rich emotional expressiveness across three modalities, models trained on it exhibit the best generalization to other corpora. We also reveal that AFEW annotations better correlate with MOSEI, MELD, and IEMOCAP and yield the best cross-corpus performance, consistent with widely-accepted basic-emotion theories.