lucidrains / phenaki-pytorch

Implementation of Phenaki Video, which uses Mask GIT to produce text guided videos of up to 2 minutes in length, in Pytorch
MIT License
740 stars 78 forks source link

Different video sizes #13

Open gmegh opened 1 year ago

gmegh commented 1 year ago

While yesterday's updates allow for all training videos to be rectangular, there is no current way to allow them to be different sizes among each other, I believe

lucidrains commented 1 year ago

@gmegh oh yes i can build that too, as long as the image dimensions are divisible by patch size (and each batch only contains the same size videos)