I have some d model renders and I want to make them look like an anime, how can I do that? I tried it like you have shown in the video but it didn't work
You probably wont need the style Controlnet like I showed here. Try using an anime style model and setting the strength low, about 45% or 50%. Maybe try the Depth or Line Art controlnets if you need it to be more accurate.
Having lived in Japan I can say you pronounced Gyoza quite well. Their delicious by the way ;) Also thanks for this info. I hadn't realized there was an update, but this is so useful and you made it exceedingly easy to get setup!
Thanks! I learned some Japanese a long time ago being an anime nerd. I had to google what gyoza actually was though. I figured it was food. I'll have to try it sometime!
Hey, just curious if you know what the difference between the "reference" and "composition" controlnets do? I've been messing around and they both seem to bring similar results.
Honestly, I'm not sure exactly. The Style, Composition, Reference and Face controlnets use different IP Adapter models. I am not sure which IP Adapter model it is using for "Reference" as opposed to "Composition". Actually, come to think of it, they could be using the same model with different settings. I wish there were better documentation about what is going on in the background.
@@IntelligentImage I did a bunch of testing and really the only thing I'm certain of is that reference tries to recreate the image. eg: If my reference image is a woman sitting in a car and I prompt something generic like "a 20 year old woman" then the results will be a woman sitting in a car. I'm guessing composition is meant to capture the look and feel as far as color, tone, shadows/highlights etc and styles is more for vector art, clay, photography, paper art etc. Much like if you were to use a1111 or comfy you can add a style that adds additional words to the prompt like "vector art, flat shading, cell shading" etc. Only an image to image version of that, idk.
I've tried switching from Krita to photoshop for painting several times and have always gone back to Krita. Photoshop wasn't originally designed for digital painting and it still shows in my opinion. Photoshop's generative fill is still superior enough for me to keep it around for the time being though since I am already paying for other Adobe programs.
If you are using the local managed server for the plugin, it should download them for you automatically or you should be able to select them and download them from the list under the local managed server in the settings.
I just uninstalled this because it used to be to basic to be useful like webforge ui or it’s more difficult be more detailed and controllable comfyui, might need to re-install but I wish they would let you host stable diffusion on another drive other than c:
I have really been enjoying Krita diffusuin for a while now. Just being able to use layers and save selections for later use blows my mind. (like)
Absolutely! I feel like I'm just scratching the surface of the possibilities.
Great video! I've been following your channel with excitement since I found it. I look forward to seeing more about the use of controlnet, etc.
😀
Thanks! I'm working on a video putting everything into an actual workflow.
Whoever integrated this they have the right idea, this is so awesome! Awesome creating tutorials for this!
This is definitely the most promising tool for direct integration of AI in digital painting. It's amazing we are relying on small developers for this!
I think the he developer is Acly and they surely deserve a massive round of applause (they do some RUclips and obviously on GitHub)
I met the same situation of adding new sampler settings but not working, thanks for explaining, and thanks to chatgpt too. Really handy!
Glad it was helpful!
@@IntelligentImage BTW, I like your format so much, focusing on how to utilize Krita for real project is most helpful. Thanks for sharing!
Thanks! I've got some more videos planned like this.
Brilliant tutorial, much indebted!
Thanks! Glad it was helpful!
I have some d model renders and I want to make them look like an anime, how can I do that? I tried it like you have shown in the video but it didn't work
You probably wont need the style Controlnet like I showed here. Try using an anime style model and setting the strength low, about 45% or 50%. Maybe try the Depth or Line Art controlnets if you need it to be more accurate.
Having lived in Japan I can say you pronounced Gyoza quite well.
Their delicious by the way ;)
Also thanks for this info. I hadn't realized there was an update, but this is so useful and you made it exceedingly easy to get setup!
Thanks! I learned some Japanese a long time ago being an anime nerd. I had to google what gyoza actually was though. I figured it was food. I'll have to try it sometime!
Great video and content mate,subbed ! Keep em coming !
Thanks, will do!
Hey, just curious if you know what the difference between the "reference" and "composition" controlnets do? I've been messing around and they both seem to bring similar results.
Honestly, I'm not sure exactly. The Style, Composition, Reference and Face controlnets use different IP Adapter models. I am not sure which IP Adapter model it is using for "Reference" as opposed to "Composition". Actually, come to think of it, they could be using the same model with different settings. I wish there were better documentation about what is going on in the background.
@@IntelligentImage I did a bunch of testing and really the only thing I'm certain of is that reference tries to recreate the image. eg: If my reference image is a woman sitting in a car and I prompt something generic like "a 20 year old woman" then the results will be a woman sitting in a car.
I'm guessing composition is meant to capture the look and feel as far as color, tone, shadows/highlights etc and styles is more for vector art, clay, photography, paper art etc. Much like if you were to use a1111 or comfy you can add a style that adds additional words to the prompt like "vector art, flat shading, cell shading" etc. Only an image to image version of that, idk.
just downloaded krita and got rid of my midj/photoshop subs. ty
I've tried switching from Krita to photoshop for painting several times and have always gone back to Krita. Photoshop wasn't originally designed for digital painting and it still shows in my opinion. Photoshop's generative fill is still superior enough for me to keep it around for the time being though since I am already paying for other Adobe programs.
Maravilloso. 🇦🇷
Thanks!
Where can I get the ControlNet models and where do I put them?
Edit: I'm using custom(local) server.
If you are using the local managed server for the plugin, it should download them for you automatically or you should be able to select them and download them from the list under the local managed server in the settings.
I just uninstalled this because it used to be to basic to be useful like webforge ui or it’s more difficult be more detailed and controllable comfyui, might need to re-install but I wish they would let you host stable diffusion on another drive other than c:
You have the option to select where it installs the local managed server. I have it on my D: drive.
This shit's crazy 💀
Word