An end to end framework with adaptive spatio-temporal attention module for human action recognition

HIGHLIGHTS

• who: Spatio-Temporal Attention Module and collaborators from the Center for Robotics, School of Control Science and Engineering, Shandong University, Jinan, China have published the Article: An End to End Framework With Adaptive Spatio-Temporal Attention Module for Human Action Recognition, in the Journal: (JOURNAL)
• what: The authors demonstrate the effectiveness of SAM and TAM for action recognition respectively. The authors design a simple yet effective CNN-based attention mechanism to extract more distinctive features at minimal additional computational cost. In SAM, the authors design two new kinds of features to describe feature map, namely . . .
  

   

  

  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.