search

tencent-ailab / IP-Adapter

The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.
Apache License 2.0
4.46k stars 289 forks source link

How to generate full body image? #360

Closed silence-tang closed 1 month ago

silence-tang commented 1 month ago

Hi, may I ask how to generate full body images? Is it able to do this by setting some special promots like "full length, full body, ..." or a ControlNet with depth image / pose image condition is needed?

andapka commented 1 month ago

I'm using with ControlNet. Tried depth maps, poses, but stopped on NormalBae maps

xiaohu2015 commented 1 month ago

you can use it with controlnet

silence-tang commented 1 month ago

Thanks! I will try controlnet.