WDMIR uses wavelet-driven video-audio fusion and cross-modal attention to achieve state-of-the-art intent recognition on MIntRec and MELD-DA.
wav2vec 2.0: A framework for self-supervised learning of speech repre- sentations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.MM 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
WDMIR: Wavelet-Driven Multimodal Intent Recognition
WDMIR uses wavelet-driven video-audio fusion and cross-modal attention to achieve state-of-the-art intent recognition on MIntRec and MELD-DA.