Get better quality with SDXL

magic-research / magic-animate

[CVPR 2024] MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model

https://showlab.github.io/magicanimate/

BSD 3-Clause "New" or "Revised" License

10.5k stars 1.08k forks source link

Open Pevernow opened 10 months ago

Pevernow commented 10 months ago

It is reported that the real images generated by SDXL have higher resolution and accuracy,

which can reduce the artifacts and errors generated and improve the similarity with the input images.

T2I can also get better results with SDXL's improved text encoder.

T0L0ve commented 10 months ago

so how to replace sd1.5 with sdxl,the text_emb size is different

johndpope commented 10 months ago

these guys released training code https://github.com/MooreThreads/Moore-AnimateAnyone/ should be straightforward to switch out base model and upgrade relevant dimensions.