---
title: "WDMIR: Wavelet-Driven Multimodal Intent Recognition"
authors:
  - "Weiyin Gong"
  - "Kai Zhang"
  - "Yanghai Zhang"
  - "Qi Liu"
  - "Xinjie Sun"
  - "Junyu Lu"
  - "Linbo Zhu"
doi: "10.24963/ijcai.2025/582"
url: "https://ijcai.org/proceedings/2025/582"
pdf: "https://ijcai.org/proceedings/2025/0582.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "5226-5234"
track: "Main Track"
keywords:
  - "Natural Language Processing: NLP: Natural language semantics"
  - "Machine Learning: ML: Multi-view learning"
  - "Natural Language Processing: NLP: Information extraction"
---

# WDMIR: Wavelet-Driven Multimodal Intent Recognition

Weiyin Gong, Kai Zhang, Yanghai Zhang, Qi Liu, Xinjie Sun, Junyu Lu, Linbo Zhu

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 5226-5234. DOI: [10.24963/ijcai.2025/582](https://doi.org/10.24963/ijcai.2025/582)

## 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.