@@top1000gadgets why would anyone want to buy a stock photo made by AI? When they could just use the AI themselves to get the exact photo they ask for, and for cheaper than buying yours?
Thank you very much for sharing this. i have been searching for controllnet models for a while. and happy to see that here. and Thanks for guiding us through what other models do...
Great job! I like that you have good focus - what are the different preproccesors and what do they do? Some other guys talk too much about irrelvant stuff - like what is control net or hiw to install it.
I really liked the way you explain things. Well done! Can you explain how to create drawings/illustrations from rough sketches. Something similar to Scribble Diffusion
thanks, but what lacks in your tutorial is mentioning which checkpoint you are using and why. Because some models are not compatible with every checkpoints. Would be better to mention.
Thank you for sharing! I would like to see what else we can do with ControlNet. Everyone uses OpenPose or Canny and depth. If you have more different examples, could you please send me a link?
I have covered most of the ControlNet models (except Tiles and Inpaint). What are your favorite ControlNets to use? I would love to learn about your workflow.
This is very helpful as I've been struggling with Control Net. I wonder if you could do a tutorial on uploading an image of a face and then transforming it with Control Net to different styles? I've been trying to do this with Canny but it keeps losing the original look of the face.
Love your content. Thank you for all the work you put out. Been having problems running some Preprocessors in ControlNet a1111 and hoping you can help me. I've tried updating, deleting and reinstalling, I've tried running SD a1111 in different browsers, redownloading the models, tried etc., and I just keep getting the following error: " urllib.error.URLError: " Anything come to mind that could help me solve this? Like I mentioned, only a few preprocessors work. For example, for "Depth," only the "depth_zoe" works, all others give the error. For "OpenPose", none of the preprocessors work. This happens both in SDXL and in 1.5 models. "Canny" on the other hand, they all work. Any ideas would be great. Thank you in advance.
Good job for the video it was clear ! But I have a question... Can we recreate with controlnet the lineart of a image ? ( Not in black background and white line, pure white background and black line ). I see that in a another video but I didn't understand. Thanks in advance
Great video and there's a lot to absorb all at once but I learned some new things. I'd love to know more about how these different techniques can be applied to batch image processing for animation, which is what I'm working on, right now. Thanks.
Thanks! Right now, I'm making videos for beginners, but I'll cover specific topics in the future. If you're working on batch image processing for animation, you can try checking out an extension called SD-CN-Animation. It's easy to use and great for Video to Video animation.
write notes that'll help you, i have the same problem of taking in info at once, once i jot it down it becomes so much easier to remember and also reference later on.
OpenPose struggles with harder poses and when hands are not perfectly visible. Make sure to have a good resolution reference. You can find poses and export OpenPose images on the website app.posemy.art/ . It works better than using reference images
If you haven't done so already, you could try increasing the strength of the controlnet slightly and select "controlnet is more important". OpenPose isn't perfect, and sometimes it doesn't generate accurately. In such cases, you might need to use inpainting or Photoshop to fix the images.
Did github remove controlnet from their repository? I finally installed Automatic 1111 and about 12 extensions but controlnet was not available. I'm a newb so maybe I'm losing my mind and not seeing it, can someone comment and tell me controlnet extension for Auto1111 is there? Anyone can comment on this question, thanks
It's the Dynamic Prompt Extension that selects a random keyword from a text file with each generation. For a brief explanation, check out my video on it th-cam.com/video/f8KUkWTaV_Q/w-d-xo.htmlsi=JrFuUYkfhpAStNkG&t=116
If you just want to extract video frames, then use FFmpeg or Shutter Encoder. If you want to save the preprocessor results for the batch frames, go to Settings > Controlnet > Directory for detected maps auto-saving > Enable 'Allow detectmap auto-saving'. In the controlnet batch process, select the extracted frames folder and process it using Sampling steps set to 1 to save time
hey man, i wish the tutorial was less theory and more practice. For every image you generated, plz give us the exact steps, prompt...etc...you're going way too fast
I really like your style of tutorial- very quick and to the point. A lot of creators would give this same amount of information in 3x the time
Can these AI arts sell stock sites? Have licence ?
@@top1000gadgets why would anyone want to buy a stock photo made by AI? When they could just use the AI themselves to get the exact photo they ask for, and for cheaper than buying yours?
Ai generators not free.@@Ascended23
great video! I was searching for a detailed explanation on all the types in control net. Good thing i came across your video!
Gosh how much something like this was needed! Excellent work!
Exactly the quick info I was looking for to get an idea of each setting.
Thank you very much for sharing this. i have been searching for controllnet models for a while. and happy to see that here.
and Thanks for guiding us through what other models do...
Bro that’s some 100k TH-cam channel quality content! Keep it up
Very good! I like how much control it gives over the images!
bro ! u did good job ,i like how u break down and it is clear to pick up ! great job
Glad you liked the video! Thanks!
Thanks for the quick and concise descriptions, appreciated.
Thank you so much!!! Precise and concise! so much value in just a few minutes 💎✨
Great visual overview of the ControlNet options!
Informative and clear explanation in short video, great video!
Man you stuff is soooo good. Short and straight to the meat. Full of value. I am really looking forward to more videos
Great video. I just subscribed. More videos describing your workflow and a little of the hows and why would be most appreciated.
Great job! I like that you have good focus - what are the different preproccesors and what do they do? Some other guys talk too much about irrelvant stuff - like what is control net or hiw to install it.
Thank you for the information.
Thank you so much for the great information! Your videos are amazingly helpful! I hope you make some more videos soon! :)
man your tutorial is sooo good, no bullshit, to the point, no cryptic steps, no useless conversation, Definitely Subs!
ขอบคุณสำหรับการสอนที่ดีเยี่ยม
Thanks for the tut
Very clear and informative, can you also make tutorial on how to make Lora
Yea, Lora training guide is on my list. Might take some time.
wow!! excellent explanation of all the options!
Wow, thanks, super useful explanations, already got the results right away
Very helpful and easy to understand. Thank you very much for the video
Excellent video. Well done and thanks! I will recommend this to beginners!
Thank you! I appreciate your support :)
Very nice, I think I am going to watch it a second time, subscribed
Thanks for the guide !
i learn a lot from this video.. thank you so much..
Muy buen video!! muchisimas gracias por el contenido
I really liked the way you explain things. Well done!
Can you explain how to create drawings/illustrations from rough sketches. Something similar to Scribble Diffusion
Thank you for The information, this video has inspired me greatly
thanks, but what lacks in your tutorial is mentioning which checkpoint you are using and why. Because some models are not compatible with every checkpoints. Would be better to mention.
Anyone has experienced that the openpose does not work at all?
Wow thank you so much for this video. Sooo helpful! Subscribed!
Excellent tutorial, thanks!
I love this video fast easy to understand no bs
Great video
Thanks a lot this was very helpful
very insightful, thank you for sharing
Glad you enjoyed it
thankyou verry much you helped me
Thank you for sharing! I would like to see what else we can do with ControlNet. Everyone uses OpenPose or Canny and depth. If you have more different examples, could you please send me a link?
I have covered most of the ControlNet models (except Tiles and Inpaint).
What are your favorite ControlNets to use? I would love to learn about your workflow.
i m repeatly watching this video to get the idea of how it can produce images ! well done
DUDE THANK YOU SO MUCH
Bro. Thank you!
very useful thank you!
🎉🎉🎉 make video about training please ❤❤❤
Well done!
This is very helpful as I've been struggling with Control Net.
I wonder if you could do a tutorial on uploading an image of a face and then transforming it with Control Net to different styles? I've been trying to do this with Canny but it keeps losing the original look of the face.
Love your content. Thank you for all the work you put out.
Been having problems running some Preprocessors in ControlNet a1111 and hoping you can help me.
I've tried updating, deleting and reinstalling, I've tried running SD a1111 in different browsers, redownloading the models, tried etc., and I just keep getting the following error:
"
urllib.error.URLError:
"
Anything come to mind that could help me solve this?
Like I mentioned, only a few preprocessors work.
For example, for "Depth," only the "depth_zoe" works, all others give the error. For "OpenPose", none of the preprocessors work. This happens both in SDXL and in 1.5 models. "Canny" on the other hand, they all work.
Any ideas would be great. Thank you in advance.
What does "Canny Low Threshold" and "Canny High Threshold" do if we raise/lower either?
Good job for the video it was clear ! But I have a question... Can we recreate with controlnet the lineart of a image ? ( Not in black background and white line, pure white background and black line ).
I see that in a another video but I didn't understand. Thanks in advance
You can re-upload the white line/black background image and choose 'invert' from the preprocessor options to get that
Great video and there's a lot to absorb all at once but I learned some new things. I'd love to know more about how these different techniques can be applied to batch image processing for animation, which is what I'm working on, right now. Thanks.
Thanks! Right now, I'm making videos for beginners, but I'll cover specific topics in the future. If you're working on batch image processing for animation, you can try checking out an extension called SD-CN-Animation. It's easy to use and great for Video to Video animation.
write notes that'll help you, i have the same problem of taking in info at once, once i jot it down it becomes so much easier to remember and also reference later on.
👍👍👍
Thanks bro..!
Can you do a deep dive into reference only? my results are all over the place so far..
When you use open pose and you click on edit. If you right click on a key point it removes it is there a way to add a key point?
All of these tutorials on controlnet skips the Tiles tab, and we still don't know what it is for.
I don't have those settings for control type. How do I get the reference option added?
can you apply a shuffle color scheme to an already made image? if so, how?
i am looking for this video thnz
could you pls explain how the inpaint global harmonious preprocessor works?
hello, the model you use for lineart is not in the download link...
i'm trying to get consistent image while using different poses.
Can i combine reference and pose ? How would that work ? I'm pretty new to this still
Use Multicontrolnet as shown at the end of the video. Choose an image with your pose for OpenPose and reference image for reference_only
When I use open pose. The resulting image doesn't have the hands correctly fixed. I did use the open pose full but still.
OpenPose struggles with harder poses and when hands are not perfectly visible. Make sure to have a good resolution reference.
You can find poses and export OpenPose images on the website app.posemy.art/ . It works better than using reference images
@@Einogen actually the open image I get in the preview is correct. But when I generate the image that's when things go south.
If you haven't done so already, you could try increasing the strength of the controlnet slightly and select "controlnet is more important". OpenPose isn't perfect, and sometimes it doesn't generate accurately. In such cases, you might need to use inpainting or Photoshop to fix the images.
@@Einogen yeah, inpaint seems to be way to go.
i'm full
Did github remove controlnet from their repository? I finally installed Automatic 1111 and about 12 extensions but controlnet was not available. I'm a newb so maybe I'm losing my mind and not seeing it, can someone comment and tell me controlnet extension for Auto1111 is there? Anyone can comment on this question, thanks
Its still there github.com/Mikubill/sd-webui-controlnet. You can copy-paste the link and click install from URL on the extensions tab.
What does 2 underscores do in your prompt? I see you used _ _ dress _ _
It's the Dynamic Prompt Extension that selects a random keyword from a text file with each generation. For a brief explanation, check out my video on it th-cam.com/video/f8KUkWTaV_Q/w-d-xo.htmlsi=JrFuUYkfhpAStNkG&t=116
@@Einogen oh cool, thanks!
I don't know what I do wrong but controlnet don't listen to me half of the times
Can I use this with EasyDiffusion?
Hello, tell me what could be the problem, I installed controlnet, but for some reason I have a control type like yours
Try updating controlnet from the extension tab.
@@Einogen thank you very much you helped me a lot
Can you tel me how to export videos frame by frame pngs via control net .
Thanks
If you just want to extract video frames, then use FFmpeg or Shutter Encoder.
If you want to save the preprocessor results for the batch frames, go to Settings > Controlnet > Directory for detected maps auto-saving > Enable 'Allow detectmap auto-saving'.
In the controlnet batch process, select the extracted frames folder and process it using Sampling steps set to 1 to save time
hello!
can someone tell me i can use this in google colab
do i need to pay for these
I dont have extension tab what do I do?
can you please give the link to where you downloaded your stable diffusion?
github.com/AUTOMATIC1111/stable-diffusion-webui Here's the link. Follow the instructions there
hey man, i wish the tutorial was less theory and more practice. For every image you generated, plz give us the exact steps, prompt...etc...you're going way too fast
that is just stealing art on another level
I did everything you did but ı dont see controlnet folder, why? 0:34
What does the "All" Control Type setting do?