Nice tutorial thank you! I saw some inconsistancies for example when you said change the jacket color it changed the style of the jacket too. That would be a problem for some work that I would want to do, as an example I would want a character to wear the same clothes in a video in many scenes so it's consistant. Anyhow, I'll play with the free version.
Also try going to 'Pro' on the left and choosing a different model. It seems to use Flux by default and it could be the training of their model causing that issue.
how do you get the model to smile or change her facial expression? like a smile or a frown, like a happy face or angry face etc.? instead of having the same face expression in all photos?
I would describe it in the prompt mate! I just tested it then to be sure, I added "big happy smile on her face" to a model from the video and it nailed it.
most of the realistic characters I create are consistent but their eyes are all blurry, rendernet seems to not be able to create clear eyes. Any advice??
@@WadeMcMaster I am of course. I have noticed the eyes seem to be even more cartoonish with the new flux mode so I switched it off. Could you make a new video on the updated Rendernet?
All good mate, try googling rendernet if you want to try. What country are you in? It could be a server or routing issue as the link works here and on my phone
I Find it all absolutely amazing and i've made a few attempts myself, getting excellent results BUT ... i have a question : Let's say that i created a good picture of a character (for example, a hyper-realistic female model); i'd like to get some DIFFERENT VERSIONs of that character with some variations (hair color, skin tone, or even ethnicity), keeping EXACTLY the SAME face, the same clothing and the same pose (OK, in the case of a change of ethnicity the face will have some changes, that's natural !) ... How can I go about doing this? It seems that artificial intelligence has a hard time repeating itself! I really need a suggestion, thanks in advance !
Hey mate! I'm glad you're having fun :) You should be able to download one of your images, then reupload into the 'facelock' area (sits above the prompt bar). That should reference the face quite closely
@@WadeMcMaster I wish there was a tool where you could generate a face and then start manipulating the facial features in 3D space, like a CAD model (but photoreal). Or use prompts to alter individual features like "make the nose thinner" or "make the eyebrows thicker" or "make the cheekbones more defined" and it would alter just that one thing, so you could basically edit an existing face without it reimagining a new face each time. Same with the body too. I guess we'll get there in the future.
The only character consistency I get are the characters 404 error! I googled, searched, used the link but; Im on Mac 10.14.6 Safari. I have pro Leonardo, Playground and I've tried them all; Seart, Tensor, Clipdrop, Stablecog, no probs, only render net!
Here's a few options which may help. I just covered OpenArtAI here which has some face features: ruclips.net/video/gTwp0jVWAgM/видео.html Otherwise I typically recommend (some are affiliate links): Midjourney (best simple image generator): www.midjourney.com Leonardo AI (Best all rounder platform with a lot of features) : app.leonardo.ai/?via=wade Dzine AI (GreatAI Art Editing Suite but has some specific Face Tools): www.dzine.ai/?via=wade
? the faces are absolutely different. I am the only one, who can see that? I also tried it with several custom pics and showed the results to the subjects, who were the models. No one has seen her/himself.....;-)
This is not the free version man you confuse me because you probably use the pro version and not the basic free i want a tutorial with the free version.
I haven't noticed that sorry! But I would say until their models improve, editing in canvas. I did see a negative prompt from Reddit user Adprevious2308: Negative Prompt: The artwork avoids the pitfalls of bad art, such as ugly and deformed eyes and faces, poorly drawn, blurry, and disfigured bodies with extra limbs and close-ups that look weird. It also avoids other common issues such as watermarking, text errors, missing fingers or digits, cropping, poor quality, and JPEG artifacts. The artwork is free of signature or watermark and avoids framing issues. The hands are not deformed, the eyes are not disfigured, and there are no extra bodies or limbs. The artwork is not blurry, out of focus, or poorly drawn, and the proportions are not bad or deformed. There are no mutations, missing limbs, or floating or disconnected limbs. The hands and neck are not malformed, and there are no extra heads or out-of-frame elements. The artwork is not low-res or disgusting and is a well-drawn, highly detailed, and beautiful rendering. Since rendernet is stable diffusion based, this is for Leonardo AI which is also stable diffusion based. So I'd try it out!
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
Nice tutorial thank you! I saw some inconsistancies for example when you said change the jacket color it changed the style of the jacket too. That would be a problem for some work that I would want to do, as an example I would want a character to wear the same clothes in a video in many scenes so it's consistant. Anyhow, I'll play with the free version.
True, I didn't consider that specifically. Thank you for the input! I hope you enjoy playing with it :)
Looks really interesting. Great to see SD functionality meets play-nice UI. 🥳
Yeah a lot of these platforms are popping up and doing a good job. I do like the face lock feature though, pretty handy :)
I click on a character I created and it doesn't allow me to generate image. It keeps asking me to select a model.
I may need to review the process, but I think you need to create the character first
Is it possible to have multiple characters together in the same image?
Oh, not 100% sure on rendernet. I'd need to investigate a bit more...
2 questions: 1. Is there also a 'Bodylock' so I can upload a face and body image? 2. Is anything censored on this site?
No sure about censorship, but last i checked there is a pose section under controlnet - would need to be a stable diffusion model
How about using consistent characters for videos?
I don't think it's a video nature anywhere yet, however, using an IMG to video tool, importing images made with a consistent character could help
Can I add the IA model to my own videos for Instagram and Facebook?
I don't see why not, a lot of people do I believe
Thanks for the helpful and very well presented content.
No problem mate, thank you for checking it out!
Why does my user interface look nothing like yours?
This video is a bit older and the design has been updated, so it is a little different
Why can't I get a picture with a sharp background, the background is always blurry. How do you arrange it?
I haven't used it for a while sorry mate, but you can try adding (small aperture)
Also try going to 'Pro' on the left and choosing a different model. It seems to use Flux by default and it could be the training of their model causing that issue.
how do you get the model to smile or change her facial expression? like a smile or a frown, like a happy face or angry face etc.? instead of having the same face expression in all photos?
I would describe it in the prompt mate! I just tested it then to be sure, I added "big happy smile on her face" to a model from the video and it nailed it.
most of the realistic characters I create are consistent but their eyes are all blurry, rendernet seems to not be able to create clear eyes. Any advice??
that's odd mate, I haven't had that issue. you could describe the eyes in your prompt a little further possibly
@@WadeMcMaster I am of course. I have noticed the eyes seem to be even more cartoonish with the new flux mode so I switched it off. Could you make a new video on the updated Rendernet?
I'll pop it onto my list but might take a while unfortunately
I just tried again, same again. I made video screen in Quicktime if you want to see.
All good mate, try googling rendernet if you want to try. What country are you in? It could be a server or routing issue as the link works here and on my phone
I Find it all absolutely amazing and i've made a few attempts myself, getting excellent results BUT ... i have a question : Let's say that i created a good picture of a character (for example, a hyper-realistic female model); i'd like to get some DIFFERENT VERSIONs of that character with some variations (hair color, skin tone, or even ethnicity), keeping EXACTLY the SAME face, the same clothing and the same pose (OK, in the case of a change of ethnicity the face will have some changes, that's natural !) ... How can I go about doing this? It seems that artificial intelligence has a hard time repeating itself! I really need a suggestion, thanks in advance !
Hey mate! I'm glad you're having fun :) You should be able to download one of your images, then reupload into the 'facelock' area (sits above the prompt bar). That should reference the face quite closely
Is there a way to change the actual face into a new face that doesn't exist IRL, or can you only use a face from a photo?
You could generate a face wjth a prompt alone, download the image and reupload to facelock
@@WadeMcMaster I wish there was a tool where you could generate a face and then start manipulating the facial features in 3D space, like a CAD model (but photoreal). Or use prompts to alter individual features like "make the nose thinner" or "make the eyebrows thicker" or "make the cheekbones more defined" and it would alter just that one thing, so you could basically edit an existing face without it reimagining a new face each time. Same with the body too. I guess we'll get there in the future.
Oh, not sure about that sorry mate. I'd love to know about that myself.
How can I delete my photo in the introducing character lab ???????
I had someone ask that question recently, I couldn't figure out how to do it unfortunately. I'd consider contacting them on discord or something
Thanks a lot …please inform me ASAP
I mean you coild consider contacting them. I may do it, but it wouldn't be soon
Error 404 Page not found...from your link
I just tried and it worked - is this for the Rendernet link?
How do I delete a Character I don't want in the free version?
Thank you in advance,
Robert
I can't see a way to do that sorry mate :/
The only character consistency I get are the characters 404 error! I googled, searched, used the link but; Im on Mac 10.14.6 Safari. I have pro Leonardo, Playground and I've tried them all; Seart, Tensor, Clipdrop, Stablecog, no probs, only render net!
That's odd mate, I'll let these guys know.
The problem is the colors, I put green hair with black t-shirt and always comes with green cloths
That sucks mate, sorry to hear. Hopefully that improves soon. I know these platforms are constantly improving.
Please how can I upload my branded t-shirt so that the AI can wear my branded clothings?
Good question! Not 100% sure sorry. If I were doing it currently I'd be photoshopping the clothing still, to know the results accurate)
need alternative
Here's a few options which may help. I just covered OpenArtAI here which has some face features: ruclips.net/video/gTwp0jVWAgM/видео.html
Otherwise I typically recommend (some are affiliate links):
Midjourney (best simple image generator):
www.midjourney.com
Leonardo AI (Best all rounder platform with a lot of features) :
app.leonardo.ai/?via=wade
Dzine AI (GreatAI Art Editing Suite but has some specific Face Tools):
www.dzine.ai/?via=wade
I hope this will be real public technology like maya and blender in 20 years time
It'll be a whole new level by then for sure!
? the faces are absolutely different. I am the only one, who can see that? I also tried it with several custom pics and showed the results to the subjects, who were the models. No one has seen her/himself.....;-)
Sorry to hear that man, it did work well for me so not sure why you're having trouble :/ wish I could be more helpful.
This is not the free version man you confuse me because you probably use the pro version and not the basic free i want a tutorial with the free version.
I don't remember paying for it lol. Maybe some features have become premium since. Either way I have no plans on doing another video sorry mate.
👋
👏
HANDS are pretty bad
do you know how to fix it on rendernet?
I haven't noticed that sorry! But I would say until their models improve, editing in canvas.
I did see a negative prompt from Reddit user Adprevious2308:
Negative Prompt:
The artwork avoids the pitfalls of bad art, such as ugly and deformed eyes and faces, poorly drawn, blurry, and disfigured bodies with extra limbs and close-ups that look weird. It also avoids other common issues such as watermarking, text errors, missing fingers or digits, cropping, poor quality, and JPEG artifacts. The artwork is free of signature or watermark and avoids framing issues. The hands are not deformed, the eyes are not disfigured, and there are no extra bodies or limbs. The artwork is not blurry, out of focus, or poorly drawn, and the proportions are not bad or deformed. There are no mutations, missing limbs, or floating or disconnected limbs. The hands and neck are not malformed, and there are no extra heads or out-of-frame elements. The artwork is not low-res or disgusting and is a well-drawn, highly detailed, and beautiful rendering.
Since rendernet is stable diffusion based, this is for Leonardo AI which is also stable diffusion based. So I'd try it out!