MIAR: Modality Interaction and Alignment Representation Fuison for Multimodal Emotion

TL;DR

The paper introduces MIAR, a novel multimodal emotion recognition method that effectively integrates and aligns language, vision, and audio modalities, improving performance and generalization across diverse datasets.

Contribution

MIAR uniquely combines feature interaction and contrastive alignment to better handle modality differences and enhance multimodal emotion recognition accuracy.

Findings

MIAR outperforms existing methods on CMU-MOSI and CMU-MOSEI datasets.

The approach improves cross-modal alignment and global feature representation.

Experimental results demonstrate superior performance over state-of-the-art techniques.

Abstract

Multimodal Emotion Recognition (MER) aims to perceive human emotions through three modes: language, vision, and audio. Previous methods primarily focused on modal fusion without adequately addressing significant distributional differences among modalities or considering their varying contributions to the task. They also lacked robust generalization capabilities across diverse textual model features, thus limiting performance in multimodal scenarios. Therefore, we propose a novel approach called Modality Interaction and Alignment Representation (MIAR). This network integrates contextual features across different modalities using a feature interaction to generate feature tokens to represent global representations of this modality extracting information from other modalities. These four tokens represent global representations of how each modality extracts information from others. MIAR aligns different modalities using contrastive learning and normalization strategies. We conduct experiments on two benchmarks: CMU-MOSI and CMU-MOSEI datasets, experimental results demonstrate the MIAR outperforms state-of-the-art MER methods.