yes I know, it's an unfortunate event that I work on the same stuff as FallenIncursio, but I think this does provide additional value, so I am still posting this.
Content included:
A - frame of 2 characters, upper character with both hands raised is best inpaint with canny (a (2).jpg); bottom character is best inpaint with depth (a (1) or a (3).jpg) or simply without controlnet, because it's such a simple pose.
B - approaching each other meme, work well with fallenincursio's lora, or work alone, both ok, adjust the control weight from 0.5 - 1.0 for more freedom of the character.
Description
FAQ
Comments (6)
can you teach me how to upload poses ?
Sure thing, in which aspect do you wish to know about? Generating the canny or depth?
@yomama123556778 after i download this, where do i save it?
@zlingerfinger first, install control net if you haven't already; if you have not, please refer to GitHub - lllyasviel/ControlNet: Let us control diffusion models! Afterwards unzip my file, simply drag the depth or canny into controlNet in stable diffusion, adjust to depth or canny in model accordingly. Remember to click none for preprocessor if you drop my images a (1) (2) (3) or b (1). It's quite complicated to explain in few words, I suggest you look up for a tutorial video.
@zlingerfinger I used the same method i mentioned in part 3.2 in my article Civitai | My humble beginner guide to work with multiple Loras for complex composition. But it's not detailed enough regarding how to use controlNet I'm afraid, just for reference.
@yomama123556778 thanks