How would you compare the image processing applied by first layer of a deep Imagenet model with old contrast normalization + ZCA whitening?
Replying to @edersantana
@edersantana 2-layer convnet with no preprocessing is much better than 1-layer convnet with ZCA whitening, so learned features are superior.
0 replies
1 retweet
3 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.