HIGHLIGHTS
SUMMARY
Jokanovic et_al7 make use of Deep Neural_Networks and stacked autoencoders to detect activities on 3 s windows, showing superior results using Deep Neural_Networks, as opposed to convential and PCA based methods. Many of these works7-12 require neural_networks on a device where prediction is done on a buffer of data, rather than handling every timestep seperately. Two different models are evaluated on both Range-Doppler (RD) maps and MD signatures: a Random Forest15 and a Convolutional Neural_Network. Vandersmissen et_al8 make use of Long Short-Term Memory (LSTM) units in a deep neural_network to recognize . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.