skip to main content
Idioma:

FGW-FER: Lightweight Facial Expression Recognition with Attention

Dinh, Huy-Hoang ; Do, Hong-Quan ; Doan, Trung-Tung ; Le, Cuong ; Bach, Ngo Xuan ; Phuong, Tu Minh ; Vu, Viet-Vu

KSII transactions on Internet and information systems, 2023-09, Vol.17 (9), p.2505-2528 [Periódico revisado por pares]

한국인터넷정보학회

Texto completo disponível

Citações Citado por
  • Título:
    FGW-FER: Lightweight Facial Expression Recognition with Attention
  • Autor: Dinh, Huy-Hoang ; Do, Hong-Quan ; Doan, Trung-Tung ; Le, Cuong ; Bach, Ngo Xuan ; Phuong, Tu Minh ; Vu, Viet-Vu
  • Assuntos: Analysis ; Attention ; Computational linguistics ; Depth-wise separable convolution ; Facial expression recognition ; Language processing ; Lightweight deep learning model ; Machine learning ; Natural language interfaces ; Neural networks ; Residual block
  • É parte de: KSII transactions on Internet and information systems, 2023-09, Vol.17 (9), p.2505-2528
  • Notas: Korean Society for Internet Information
    KISTI1.1003/JNL.JAKO202330043222946
  • Descrição: The field of facial expression recognition (FER) has been actively researched to improve human-computer interaction. In recent years, deep learning techniques have gained popularity for addressing FER, with numerous studies proposing end-to-end frameworks that stack or widen significant convolutional neural network layers. While this has led to improved performance, it has also resulted in larger model sizes and longer inference times. To overcome this challenge, our work introduces a novel lightweight model architecture. The architecture incorporates three key factors: Depth-wise Separable Convolution, Residual Block, and Attention Modules. By doing so, we aim to strike a balance between model size, inference speed, and accuracy in FER tasks. Through extensive experimentation on popular benchmark FER datasets, our proposed method has demonstrated promising results. Notably, it stands out due to its substantial reduction in parameter count and faster inference time, while maintaining accuracy levels comparable to other lightweight models discussed in the existing literature.
  • Editor: 한국인터넷정보학회
  • Idioma: Coreano;Inglês
Voltar para lista de resultados
Ir para página anteriorAnterior Resultado  9 AvançarIr para próxima página

  • Realização: ABCD-USP USP   Logos de Redes Sociais: Freepik

Buscando em bases de dados remotas. Favor aguardar.