Video example?

[dreamingtulpa]

Could you provide an example on how to use this method for video inference?

[SoftologyPro]

Yes, please do. How is the first animated gif created?
Also "ScaleCrafter is capable of generating images with resolution of 4096 x 4096 and results with resolution of 2048 x 1152 based on pre-trained diffusion models on a lower resolution." is results a typo for videos?

[YingqingHe]

Also "ScaleCrafter is capable of generating images with resolution of 4096 x 4096 and results with resolution of 2048 x 1152 based on pre-trained diffusion models on a lower resolution." is results a typo for videos?

Thanks! We have fixed it.

[SoftologyPro]

Thanks! We have fixed it.

How do we create videos?

[JCBrouwer]

This method only overrides the configurations of some of the convolutions in the UNet as far as I can tell.

Essentially any video synthesis algorithm that relies on the stable diffusion UNet should be compatible, you "just" have to splice in the inflation/dilation code from lines 197-258 to your video pipeline.

[SoftologyPro]

Hopefully one of the devs gives us a working script. The movie is their first example after all. Surely we don't have to use some other system and splice in other code etc?

[YingqingHe]

Yes, we are preparing this part. Please stay tuned

[IIIIIIIllllllllIIIII]

+1 for video generation code

[zhanghongyong123456]

+1 for video generation code

+1 for video generation code

[tommy19970714]

+1 for video generation code

[zhw-zhang]

Hello, thanks for your contribution, I'm also waiting for this part, and I would like to ask when will this code be released.

[HuaizeLiu]

+1 for video generation code

[SoftologyPro]

Yes, we are preparing this part. Please stay tuned

5 months later. Any update?

[lovejing0306]

10 months later. Any update?