Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture

The development of emotionally intelligent computers depends on emotion recognition based on richer multimodal inputs, such as text, speech, and visual cues, as multiple modalities complement one another. The effectiveness of complex relationships between modalities for emotion recognition has been...

Full description

Saved in:
Bibliographic Details
Main Authors: Hajar Filali, Chafik Boulealam, Khalid El Fazazy, Adnane Mohamed Mahraz, Hamid Tairi, Jamal Riffi
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Information
Subjects:
Online Access:https://www.mdpi.com/2078-2489/16/1/40
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832588353677033472
author Hajar Filali
Chafik Boulealam
Khalid El Fazazy
Adnane Mohamed Mahraz
Hamid Tairi
Jamal Riffi
author_facet Hajar Filali
Chafik Boulealam
Khalid El Fazazy
Adnane Mohamed Mahraz
Hamid Tairi
Jamal Riffi
author_sort Hajar Filali
collection DOAJ
description The development of emotionally intelligent computers depends on emotion recognition based on richer multimodal inputs, such as text, speech, and visual cues, as multiple modalities complement one another. The effectiveness of complex relationships between modalities for emotion recognition has been demonstrated, but these relationships are still largely unexplored. Various fusion mechanisms using simply concatenated information have been the mainstay of previous research in learning multimodal representations for emotion classification, rather than fully utilizing the benefits of deep learning. In this paper, a unique deep multimodal emotion model is proposed, which uses the meaningful neural network to learn meaningful multimodal representations while classifying data. Specifically, the proposed model concatenates multimodality inputs using a graph convolutional network to extract acoustic modality, a capsule network to generate the textual modality, and vision transformer to acquire the visual modality. Despite the effectiveness of MNN, we have used it as a methodological innovation that will be fed with the previously generated vector parameters to produce better predictive results. Our suggested approach for more accurate multimodal emotion recognition has been shown through extensive examinations, producing state-of-the-art results with accuracies of 69% and 56% on two public datasets, MELD and MOSEI, respectively.
format Article
id doaj-art-80e6ae9781ae45fc8b2787461092b46c
institution Kabale University
issn 2078-2489
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Information
spelling doaj-art-80e6ae9781ae45fc8b2787461092b46c2025-01-24T13:35:15ZengMDPI AGInformation2078-24892025-01-011614010.3390/info16010040Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer ArchitectureHajar Filali0Chafik Boulealam1Khalid El Fazazy2Adnane Mohamed Mahraz3Hamid Tairi4Jamal Riffi5LISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoLISAC, Department of Computer Science, Faculty of Science Dhar El Mahraz, Sidi Mohamed Ben Abdellah University, Fez 30000, MoroccoThe development of emotionally intelligent computers depends on emotion recognition based on richer multimodal inputs, such as text, speech, and visual cues, as multiple modalities complement one another. The effectiveness of complex relationships between modalities for emotion recognition has been demonstrated, but these relationships are still largely unexplored. Various fusion mechanisms using simply concatenated information have been the mainstay of previous research in learning multimodal representations for emotion classification, rather than fully utilizing the benefits of deep learning. In this paper, a unique deep multimodal emotion model is proposed, which uses the meaningful neural network to learn meaningful multimodal representations while classifying data. Specifically, the proposed model concatenates multimodality inputs using a graph convolutional network to extract acoustic modality, a capsule network to generate the textual modality, and vision transformer to acquire the visual modality. Despite the effectiveness of MNN, we have used it as a methodological innovation that will be fed with the previously generated vector parameters to produce better predictive results. Our suggested approach for more accurate multimodal emotion recognition has been shown through extensive examinations, producing state-of-the-art results with accuracies of 69% and 56% on two public datasets, MELD and MOSEI, respectively.https://www.mdpi.com/2078-2489/16/1/40emotion recognitiondeep learninggraph convolutional networkcapsule networkvision transformermeaningful neural network (MNN)
spellingShingle Hajar Filali
Chafik Boulealam
Khalid El Fazazy
Adnane Mohamed Mahraz
Hamid Tairi
Jamal Riffi
Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
Information
emotion recognition
deep learning
graph convolutional network
capsule network
vision transformer
meaningful neural network (MNN)
title Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
title_full Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
title_fullStr Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
title_full_unstemmed Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
title_short Meaningful Multimodal Emotion Recognition Based on Capsule Graph Transformer Architecture
title_sort meaningful multimodal emotion recognition based on capsule graph transformer architecture
topic emotion recognition
deep learning
graph convolutional network
capsule network
vision transformer
meaningful neural network (MNN)
url https://www.mdpi.com/2078-2489/16/1/40
work_keys_str_mv AT hajarfilali meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture
AT chafikboulealam meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture
AT khalidelfazazy meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture
AT adnanemohamedmahraz meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture
AT hamidtairi meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture
AT jamalriffi meaningfulmultimodalemotionrecognitionbasedoncapsulegraphtransformerarchitecture