Suggestion: Init Image and LORA Support

jjihwan / FIFO-Diffusion_public

Official implementation of FIFO-Diffusion

https://jjihwan.github.io

277 stars 17 forks source link

Suggestion: Init Image and LORA Support #7

Closed Gitterman69 closed 3 weeks ago

Gitterman69 commented 1 month ago

I would like to suggest adding support for Init Image and LORA (Low-Rank Adaptation) in the txt2video AI tool. These features would enhance the tool’s flexibility and capability, allowing users to create more refined and controlled video outputs.

jjihwan commented 1 month ago

I totally agree, but I don't have enough time to implement them right now 🥲. Supporting LoRA, in particular, depends on the baseline model implementation, which in our case is VideoCrafter2. However, VC2 does not officially support LoRA yet. If you succeed in implementing it, we would welcome a pull request :)