| Texto completo | |
| Autor(es): Mostrar menos - |
Tacon, Hemerson
;
Brito, Andre S.
;
Chaves, Hugo L.
;
Vieira, Marcelo Bernardes
;
Villela, Saulo Moraes
;
Maia, Helena de Almeida
;
Concha, Darwin Ttito
;
Pedrini, Helio
;
Misra, S
;
Gervasi, O
;
Murgante, B
;
Stankova, E
;
Korkhov, V
;
Torre, C
;
Rocha, AMAC
;
Taniar, D
;
Apduhan, BO
;
Tarantino, E
Número total de Autores: 18
|
| Tipo de documento: | Artigo Científico |
| Fonte: | COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2019, PT I; v. 11619, p. 16-pg., 2019-01-01. |
| Resumo | |
Despite the expressive progress of deep learning models on the image classification task, they still need enhancement for efficient human action recognition. One way to achieve such gain is to augment the existing datasets. With this goal, we propose the usage of multiple Visual Rhythm crops, symmetrically extended in time and separated by a fixed stride. The symmetric extension preserves the video frame rate, which is crucial to not distort actions. The crops provide a 2D representation of the video volume matching the fixed input size of the 2D Convolutional Neural Network (CNN) employed. In addition, multiple crops with stride guarantee coverage of the entire video. Aiming to evaluate our method, a multi-stream strategy combining RGB and Optical Flow information is extended to include the Visual Rhythm. Accuracy rates fairly close to the state-of-the-art were obtained from the experiments with our method on the challenging UCF101 and HMDB51 datasets. (AU) | |
| Processo FAPESP: | 17/09160-1 - Reconhecimento de Ações Humanas em Vídeos |
| Beneficiário: | Helena de Almeida Maia |
| Modalidade de apoio: | Bolsas no Brasil - Doutorado |
| Processo FAPESP: | 17/12646-3 - Déjà vu: coerência temporal, espacial e de caracterização de dados heterogêneos para análise e interpretação de integridade |
| Beneficiário: | Anderson de Rezende Rocha |
| Modalidade de apoio: | Auxílio à Pesquisa - Temático |