search

nihaomiao / CVPR23_LFDM

The pytorch implementation of our CVPR 2023 paper "Conditional Image-to-Video Generation with Latent Flow Diffusion Models"
BSD 2-Clause "Simplified" License
432 stars 43 forks source link

about checkpoint #15

Closed ymmshi closed 11 months ago

ymmshi commented 11 months ago

Thanks for your great work! Where can I get the pretrained Taichi/VOX checkpoint? And are the pretrained model very important for LFAE?

image
nihaomiao commented 11 months ago

Hi, @ymmshi, thanks a lot for your interest in our work! We initialize our LFAE with the checkpoint provided in the github of MRAA. I think a good initialization will help accelerate the training of LFAE.

ymmshi commented 11 months ago

Hi Ni, thanks for your quick reply! It seems the snap's repo is over data quota, and I cannot download the checkpoints. Could you please provide your bak one if there is?

image
nihaomiao commented 11 months ago

Hi, @ymmshi, this https://github.com/snap-research/articulated-animation/issues/41#issuecomment-901416664 may solve your problem.