search

omerbt / MultiDiffusion

Official Pytorch Implementation for "MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation" presenting "MultiDiffusion" (ICML 2023)
https://multidiffusion.github.io/
979 stars 57 forks source link

Problem of Formula Derivation #29

Open hxy-123-coder opened 9 months ago

hxy-123-coder commented 9 months ago

Hello!Thank you for your great work! But I have problem on formula derivation. I met difficulties when trying to derivatie equation 3 to equation 5. image image Could you please help me and show the process. Thanks a lot!

charchit7 commented 7 months ago

Hey @omerbt @lioryariv Paper is well written and this equation make sense too. That we are following the averaging of the whole operation so that the generation is consistent. But could you please explain the derivation as mentioned by @hxy-123-coder. It will help get intuition more. Why are we doing image space inverse F^1?

Thank you!

oscarwooberry commented 1 month ago

Hey @omerbt @lioryariv Paper is well written and this equation make sense too. That we are following the averaging of the whole operation so that the generation is consistent. But could you please explain the derivation as mentioned by @hxy-123-coder. It will help get intuition more. Why are we doing image space inverse F^1?

Thank you!

Same here. I'm just wondering if you have any new insight on this.