Accurate step count with generalized and personalized deep learning on accelerometer data

HIGHLIGHTS

• who: Long Luu et al. from the The public dataset was collected at Clemson University and consists of , subjects performing three different activities (regular walking, semi-regular walking and unstructured activity) while wearing Shimmer, (Shimmer, Dublin, Ireland) devices at the hip, ankle and wrist [39]For this study, we only used the regular walking activity, in which subjects walked at a comfortable pace around a building for approximately , min. The whole walking duration was recorded with an iPhone (Apple, Cupertino, CA, USA) for subsequent manual annotation. The annotation indicates whether each time point corresponded to a left . . .
  

   

  

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

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

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