search

RedAIGC / StoryMaker

StoryMaker: Towards consistent characters in text-to-image generation
411 stars 35 forks source link

how do apply openpose when do demo_swapcloth() #19

Open tianqyun111 opened 2 days ago

tianqyun111 commented 2 days ago

i found controlnet_conditioning_scale in the pipeline_sdxl_storymaker.py is not for openpose_controlnet,how can i solve it ?

RedAIGC commented 2 days ago

controlnet_conditioning_scale

Hi, we don't use controlnet during inference by default, you can ignore controlnet_conditioning_scale.

tianqyun111 commented 2 days ago

controlnet_conditioning_scale

Hi, we don't use controlnet during inference by default, you can ignore controlnet_conditioning_scale.

Thank you for your reply. If I want to use ControlNet for the generation of clothswap, do I need to modify the pipeline myself? Can you give me some guidance

RedAIGC commented 23 hours ago

controlnet_conditioning_scale

Hi, we don't use controlnet during inference by default, you can ignore controlnet_conditioning_scale.

Thank you for your reply. If I want to use ControlNet for the generation of clothswap, do I need to modify the pipeline myself? Can you give me some guidance

That's simple, you can use the StableDiffusionXLControlNetPipeline and open-source pose controlnet model to generate your desired pose images.