Sign in

End-to-end Learning of Compressible Features

By Saurabh Singh and others
Pre-trained convolutional neural networks (CNNs) are powerful off-the-shelf feature generators and have been shown to perform very well on a variety of tasks. Unfortunately, the generated features are high dimensional and expensive to store: potentially hundreds of thousands of floats per example when processing videos. Traditional entropy based lossless compression... Show more
July 23, 2020
=
0
Loading PDF…
Loading full text...
Similar articles
Loading recommendations...
=
0
x1
End-to-end Learning of Compressible Features
Click on play to start listening