search

roimehrez / contextualLoss

The Contextual Loss
http://cgm.technion.ac.il/Computer-Graphics-Multimedia/Software/Contextual/
490 stars 78 forks source link

Input size #13

Open fdbtrs opened 6 years ago

fdbtrs commented 6 years ago

Hi Is it possible to train the network with original VGG input size 224*224 Thanks

roimehrez commented 6 years ago

The CRN architecture works well for power of 2 since the layers size are 4--8--16--32--64--128--256 You can change the network to a different one and feed any size you wish. You can use 256 over none for the CRN

Why do you need 224*224?

fdbtrs commented 6 years ago

Hi Thanks for reply. I need to validate the result against another vgg model without resizing the generated message. Mainly to check whether the model has preserved the identity or not.