HOW TO MAKE BEAUTIFUL STABLE DIFFUSION IMAGES | Negative Prompts
Вставка
- Опубліковано 24 сер 2024
- A lot of people are struggling with generating AI art to their likings on a local machine, not Midjourney or DALL-E. Negative prompts are super important for us open-source AI users. In this video, I go over what works well for me and what you could be doing wrong with your AI art generation.
Protogen x3.4 Photorealism: civitai.com/mo...
My negative prompt for portraits:
canvas frame, cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)),((close up)),((b&w)), weird colors, blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), Photoshop, video game, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, 3d render
Make sure to like and subscribe!
blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers),
what is name this site??
Excellent video. Absolutely agree that there's artistry to creating AI images. Too many people treat AI as though there's no work involved on the end user's part. Prompt Engineering, however, is a science and art in and of itself. Thanks for the helpful info here.
Thanks for the kind words, Steve! It will definitely be interesting to see how the publics view of AI art shifts as they come to realize the 'engineering' side of it. Also, checked out your Blender tuts and I'm actually going to be using those and sending them to a friend of mine! Good stuff!
Even with AI its hard to make picture, because you still need skills for prompting
Awesome. Thank you very much.
Thanks for being a continued supporter 428!
Very useful, thanks.
Glad you found it useful, thanks for the sub!
Another really nice video! Quick question, after a while of generating images i started getting 2 women consistently in every generated image. Did you get that too? Cheers!
could you do a tutorial on how to download and use the stable diffusion ui that you are using
this video shows you how to set up the same gui (remove spaces) [youtube. com/ watch?v= onfqymhBmt8]
you earn my subs thank you for all the tips as beginner to SD.
how are you using it locally?
Are you writing a book or generating a picture? 😅
I’m full into it too. The negative prompts are something hard to understand.
Haha it feel that way sometimes, open-source is cool but sometimes requires a bit more work
@@binks_live the good part is that you are much more sure of the results, while in many closed ai videos I see people discussing in the comments, they don't know for sure what their prompts do
where can i dowload ?
how can i get to this site? I use mac os and I can't find it at all.
I wish I could help but I don't have access to an Apple machine to test. Could you tell me some more about the problem?
So many cut and paste negative prompts and prompt videos out there. This video would have been much better if you walked through how best to build up a negative prompt alongside building up a prompt
How to install this model?
You can download it from civitai.com and install it into your 'models' folder under 'stable-diffusion'!
what is name this site
What is the ui you are using and where do you download it?
u ever find out i got the same question
CFG scale, not "config" but classifier-free guidance
protogen is not on the original license... sorry but no!
The SD license allows for redist. of custom models as long as the original model from which it came (SD 1.5) is stated as well as whatever modifications that have been made are clearly stated as well. Here's the link: huggingface.co/spaces/CompVis/stable-diffusion-license
@@binks_live you are not taking only the license from sd but from dreamlike...
EL MODELO DE QGO-MK8 , es mas realista, hiperrealista diría, algo fuera de lo común.😮😳😳😳, no se puede diferenciar si es verdadera o generada por AI..
¡Sí, es casi como si alguien hubiera tomado una fotografía real!
@@binks_live PERO estuve averiguando y resulto tener virus en su modelo. creo que es fake. de tener virus si lo tiene pero ya lo bloquearon en Huggi..