📝 Publications

TiCAL:Typicality-Based Consistency-Aware Learning for Multimodal Emotion Recognition
Wen Yin, Siyu Zhan, Cencen Liu, XIN Hu, Guiduo Duan, Xiurui Xie, Yuan-Fang Li, Tao He

• We propose TiCAL, a novel framework that performs dynamic multi-stage fusion by leveraging inter-modal consistency and unimodal typicality, mimicking human-like emotion perception.
• TiCAL assess unimodal emotional pseudo-labels and its typicality, then defined the modal consistency quantification matrix, enabling robust and accurate MER.

Knowledge-Aligned Counterfactual-Enhancement Diffusion Perception for Unsupervised Cross-Domain Visual Emotion Recognition
Wen Yin, Yong Wang , Guiduo Duan, Dongyang Zhang, Xin Hu, Yuan-Fang Li, Tao He

• We introduce Unsupervised Cross-Domain Visual Emotion Recognition (UCDVER).
• we propose a Knowledge-aligned Counterfactual enhancement Diffusion Perception (KCDP) framework to learn domain-agnostic knowledge across diverse emotion domains.