search

danier97 / LDMVFI

[AAAI'2024] "LDMVFI: Video Frame Interpolation with Latent Diffusion Models", Duolikun Danier, Fan Zhang, David Bull
MIT License
138 stars 15 forks source link

Pretrained model for autoencoder #4

Closed Ren-creater closed 1 year ago

Ren-creater commented 1 year ago

Hi!

It seems like you have only uploaded the pretrained model for the denoising UNet. Could you please provide the pretrained model for the autoencoder (VQ-FIGAN)? Thank you.

danier97 commented 1 year ago

Hi!

The checkpoint I uploaded actually contains the weights of both the autoencoder and the UNet.

If you load the checkpoint file (using torch.load()) and check the keys of the state_dict, you can see the parameters of the autoencoder (i.e. first_state_model.*) are also there.

I hope this helps.

Thanks.