vunet icon indicating copy to clipboard operation
vunet copied to clipboard

Published 20 hours ago • verified publisher icon CompVis

shoes dataset

Open mxkai opened this issue 5 years ago • 1 comments

Can you tell me how to train on the edges2shoes dataset

mxkai avatar Sep 09 '19 00:09 mxkai

I'm also curious about training for more general pix2pix applications, like the shoes and handbags examples on the compvis site—specifically, cases that don't use a human figure as the conditional input.

More specifically, I'm actually interested in distorting/reshaping specific images—so I would provide both a "sketch" and an image, but the sketch wouldn't be a list of joint coordinates.

jbmaxwell avatar Sep 10 '19 12:09 jbmaxwell