DeepFaceLab icon indicating copy to clipboard operation
DeepFaceLab copied to clipboard

xseg only add , 'xseg_mask': array([[137],[ 80],[ 78],...,[ 66],[ 96],[130]], dtype=uint8)}) to metadata

Open onwan946 opened this issue 3 years ago • 1 comments

Im finally realize to make meta.dat readable i simply have to pickle it. but every single jpg file that have been xseg applied only add this exactly 'xseg_mask': array([[137],[ 80],[ 78],...,[ 66],[ 96],[130]], dtype=uint8) to it dictionary I even remove xseg and conpare with the xseg one and the only different is just that.All landmark point is the same. This doesn't make any sense.

onwan946 avatar Nov 04 '22 14:11 onwan946

apparently it some kind of length limit.so i save it as json file instead

onwan946 avatar Nov 04 '22 16:11 onwan946

Issue solved / already answered (or it seems like user error), please close it.

joolstorrentecalo avatar Jun 08 '23 23:06 joolstorrentecalo