Multi-scale and attention enhanced graph convolution network for skeleton-based violence action recognition

HIGHLIGHTS

• who: Huaigang Yang from the Jiaotong University, China have published the paper: Multi-scale and attention enhanced graph convolution network for skeleton-based violence action recognition, in the Journal: (JOURNAL)
• what: In the spatial space, the authors design a multi-scale spatial GCN with a fused channel attention mechanism to extract spatial information and the correlation features between channels. Based on the success of these models, the authors design a multiscale attention spatial graph convolution network to learn spatial features from skeleton sequences, as shown in Figure 3. The authors compare the models on the . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.