WDMIR: Wavelet-Driven Multimodal Intent Recognition

TL;DR

This paper introduces a wavelet-based multimodal intent recognition framework that improves understanding of user intentions by analyzing non-verbal cues in the frequency domain, achieving state-of-the-art accuracy.

Contribution

The novel wavelet-driven fusion module and cross-modal interaction mechanism enhance multimodal intent recognition by effectively integrating verbal and non-verbal information in the frequency domain.

Findings

Achieved 1.13% higher accuracy than previous methods.

Wavelet-driven fusion significantly improves semantic extraction from non-verbal cues.

Ablation studies confirm the effectiveness of the proposed modules.

Abstract

Multimodal intent recognition (MIR) seeks to accurately interpret user intentions by integrating verbal and non-verbal information across video, audio and text modalities. While existing approaches prioritize text analysis, they often overlook the rich semantic content embedded in non-verbal cues. This paper presents a novel Wavelet-Driven Multimodal Intent Recognition(WDMIR) framework that enhances intent understanding through frequency-domain analysis of non-verbal information. To be more specific, we propose: (1) a wavelet-driven fusion module that performs synchronized decomposition and integration of video-audio features in the frequency domain, enabling fine-grained analysis of temporal dynamics; (2) a cross-modal interaction mechanism that facilitates progressive feature enhancement from bimodal to trimodal integration, effectively bridging the semantic gap between verbal and non-verbal information. Extensive experiments on MIntRec demonstrate that our approach achieves state-of-the-art performance, surpassing previous methods by 1.13% on accuracy. Ablation studies further verify that the wavelet-driven fusion module significantly improves the extraction of semantic information from non-verbal sources, with a 0.41% increase in recognition accuracy when analyzing subtle emotional cues.