Inpainting skin texture under the control of Controlnet tile and SegmentAnything, through a checkpoint of realistic skin texture.
通过真实皮肤质感的checkpoint，在Controlnet tile和SegmentAnything控制下重绘肤质
Recommended to use the realistic texture checkpoint epiCRealism - Natural Sin RC1 VAE | Stable Diffusion Checkpoint | Civitai
推荐使用真实质感checkpoint epiCRealism - Natural Sin RC1 VAE | Stable Diffusion Checkpoint | Civitai
The main model can use the SDXL checkpoint. This process is used instead of directly using the realistic texture lora because it achieves better and more controllable effects
主模型可以使用SDXL的checkpoint。使用这个流程而不是直接使用真实质感lora是因为获取更好更可控的效果
Description
FAQ
Comments (14)
好好看！！方便问下大大有在其他平台上传嘛
👍,我加载json文件之后有很多node找不到怎么办呢
安装ComfyUI-Manager 然后c菜单里Install Missing Custom Nodes
@fishdong1107 感谢回答，现在comfyUI源码报错了。UP有需要吗？改怎么解决呢。
ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
File "/ai/ComfyUI/execution.py", line 153, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "/ai/ComfyUI/execution.py", line 83, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "/ai/ComfyUI/execution.py", line 76, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "/ai/ComfyUI/custom_nodes/efficiency-nodes-comfyui/efficiency_nodes.py", line 104, in efficientloader
model, clip = load_lora(lora_params, ckpt_name, my_unique_id, cache=lora_cache, ckpt_cache=ckpt_cache, cache_overwrite=True)
File "/ai/ComfyUI/custom_nodes/efficiency-nodes-comfyui/tsc_utils.py", line 319, in load_lora
lora_model, lora_clip = recursive_load_lora(lora_params, ckpt, clip, id, ckpt_cache, cache_overwrite, folder_paths)
File "/ai/ComfyUI/custom_nodes/efficiency-nodes-comfyui/tsc_utils.py", line 310, in recursive_load_lora
lora_model, lora_clip = load_lora_for_models_tsc(ckpt, clip, lora_path, strength_model, strength_clip)
File "/ai/ComfyUI/custom_nodes/efficiency-nodes-comfyui/tsc_sd.py", line 122, in load_lora_for_models_tsc
key_map = model_lora_keys_unet(model.model)
NameError: name 'model_lora_keys_unet' is not defined
@miohu 今天efficiency-nodes-comfyui 进行了大更新，所以很多节点出错了，我重新更新了workflow。重新下载再试试
not sure if this is for SDXL 1.0 or SD 1.5.
I am using a combination of SDXL and SD1.5. The Base Model uses SDXL, while the inpaint model uses SD1.5, because I want to use CNet Tile, which is not yet in SDXL
What package did you use for the grounding dino model in your workflow?
Where is dino model? Are you hiding it because it's your treasure?
Thanks for the workflow,
what is the solution for the second part:
Error occurred when executing ControlNetApply: 'NoneType' object has no attribute 'copy'
为啥不叫洁面乳或者卸妆水:)
IM getting this error message: Ksampler: mat1 and mat2 shapes cannot be multiplied (154x2048 and 768x320)