Multimodal hand gesture recognition combining temporal and pose information based on CNN descriptors and histogram of cumulative magnitudes.

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.creatorEscobedo Cárdenas, Edwin Jonathan-
Autor(es): dc.creatorCámara Chávez, Guillermo-
Data de aceite: dc.date.accessioned2025-08-21T15:37:08Z-
Data de disponibilização: dc.date.available2025-08-21T15:37:08Z-
Data de envio: dc.date.issued2022-02-06-
Data de envio: dc.date.issued2022-02-06-
Data de envio: dc.date.issued2019-
Fonte completa do material: dc.identifierhttp://www.repositorio.ufop.br/jspui/handle/123456789/14456-
Fonte completa do material: dc.identifierhttps://www.sciencedirect.com/science/article/abs/pii/S1047320320300225-
Fonte completa do material: dc.identifierhttps://doi.org/10.1016/j.jvcir.2020.102772-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/capes/1019887-
Descrição: dc.descriptionIn this paper, we present a new approach for dynamic hand gesture recognition. Our goal is to integrate spatiotemporal features extracted from multimodal data captured by the Kinect sensor. In case the skeleton data is not provided, we apply a novel skeleton estimation method to compute temporal features. Furthermore, we introduce an effective method to extract a fixed number of keyframes to reduce the processing time. To extract pose features from RGB-D data, we take advantage of two different approaches: (1) Convolutional Neural Networks and (2) Histogram of Cumulative Magnitudes. We test different integration methods to fuse the extracted spatiotemporal features to boost recognition performance in a linear SVM classifier. Extensive experiments prove the effectiveness and feasibility of the proposed framework for hand gesture recognition.-
Formato: dc.formatapplication/pdf-
Idioma: dc.languageen-
Direitos: dc.rightsrestrito-
Palavras-chave: dc.subjectSpherical coordinates-
Palavras-chave: dc.subjectKeyframe extraction-
Palavras-chave: dc.subjectConvolucional neuronal networks-
Palavras-chave: dc.subjectFusion schemes-
Título: dc.titleMultimodal hand gesture recognition combining temporal and pose information based on CNN descriptors and histogram of cumulative magnitudes.-
Aparece nas coleções:Repositório Institucional - UFOP

Não existem arquivos associados a este item.