Thanks for doing these videos. I like your different approaches and it helps me learn this great software.

@MasterDiffusion

7 ай бұрын

You're very welcome!

Thanks , all the examples were image to image , do you get good results with plain text to image as well?

@MasterDiffusion

7 ай бұрын

Yes, absolutely. The video was about image to image. Text to image gives instant results but you have little control of the outcome.

hello there, love your work and ideas. but i have to ask why use face swap and pyrecanny? also what is CBDS? thanks in advance

@MasterDiffusion

6 ай бұрын

I guess, that’s kind of hard to explain other than it is a hack. It was not designed for doing things outside of swapping faces, but with experimenting I have found out that it has many more uses. You can make a coloring book easily without doing this. But this is a fast and reliable way of getting the exact picture that you want. Why does it work? I’m honestly not sure, but half of my time is spent just experimenting trying to figure out things that work.

@xenoroid9580

6 ай бұрын

@@MasterDiffusion that's interesting, so some tools in the program can another potential uses other than the intended. With amount of settings in the program there can potentially thousands of workflows depending on the target, this can make room for true experts to rise in the program and can achieve results better and faster than others.

@MasterDiffusion

6 ай бұрын

@@xenoroid9580 most certainly. That’s kind of my progression. I am looking for faster results with a higher degree of quality. Not always attainable though. 😂

I don't know what i'm doing wrong, but it does not work at all for me. What is the model you are using please?

@MasterDiffusion

7 ай бұрын

The default juggernaut that comes with it. Make sure you look close at my set up. No extra tags with checks. Make sure face swap is on the left and p-canny on the right. No extra Lora except the default at .1

@xellostube

6 ай бұрын

totally not working for me either

Do we have the rights to use images generated by fooocus for commercial use , for example on Amazon kdp ?

@MasterDiffusion

6 ай бұрын

Technically this is a very gray area because Ai is trained on tons of actual art and photos. The art is original but it’s made from using the actual art it’s programmed with. So with that said there is a lot of disagreement over this. But, the AI you generate has no real way of being traced to anyone’s art. You will find as many opinions and rules on this I don’t think there is a real answer yet. Sorry I’m not more help. It really becomes your choice. Try reading this allthingsai.com/article/can-you-sell-ai-generated-art-do-s-and-don-ts#:~:text=Yes%2C%20you%20can%20generally%20use,things%20to%20keep%20in%20mind.&text=AI%2Dgenerated%20art%20is%20protected,permission%20from%20the%20copyright%20holder.

Hi can you make two videos please 1) how to use prompts from a file ( automatic 1111 does it ) there is a wildcards feature in focus but I need it to read it from text file 2) how to make more images instead of being limited to 32 - may be there is a feature in advance settings Thank you

@MasterDiffusion

7 ай бұрын

I Wii sure look into it

Think there are also Lora‘s on civit

@MasterDiffusion

7 ай бұрын

There are. This was how to do it with the default

@speedracer4878

7 ай бұрын

Got any recommendations? For line art loras. btw, awesome channel :)

@MasterDiffusion

7 ай бұрын

@@speedracer4878 thanks a lot. No. I don’t do it often. I just knew how to do it from the default setup so I shared. Go to civitai and search for it. I know they have several. Just look at the reviews