Part of my facial expressions series. All of these LoRAs are designed to mesh well with each other and have low impact on your image framing.
People on cloud nine. Smiles, laughs, etc. Did this one last because SD generally can do this fairly well, but similar to shocked, it's not great at injecting the expression into more complex compositions and mixing it with other expressions.
This one sets up a full 6 prototypical expressions that can be remixed into a pretty wide palette.
Definitely has a fair amount of unintended style transfer (trends towards high saturation film-grain professional style photos) I still need to train out. Probably my last fresh one for a little bit before I do some clean up of the expression models to clear out some overfitting before trying to make a combined model.
Starter prompts:
<subject>, happy
<subject>, <context> BREAK happy expression
Description
FAQ
Comments (6)
That's a really happy face.
Joker toxin happy?
Ha! Nah, brah, just normal happy 😂. As you can tell from the combination chart, though, adding happy + doubling down the current v1 of scared might get you close.
@thegipper I was thinking mix happy with grimacing....in face I'll try that now!
Your expressions Loras are by far the most impressive here. Maybe consider sharing a detailed (just the essential for people who already know what training is) workflow. As it could help raise the quality of some other really bad mood/expression Loras.
Thanks for the hard work. Really cool to be able to play with expressions without fucking up hours of inpaint.
A big thanks !
@thegipper Yes, I would very much like to know how you trained your LoRAs. I'm experienced with training people and styles but I had no idea that it was possible to train facial expressions. I downloaded all of your LoRAs and they all worked very well with a (photo-realistic) character I trained. My intention is to render this character with a variety of facial expressions, add them to its dataset, and train again with those expressions specified. Then I tried to use your LoRAs to manifest a deep frown and was not able to do it. So I decided to train my own facial expression LoRA by collecting photos of that feature this type of frown. But my training failed. I think it's because I did not write proper captions. It would be good to know how you trained these. Otherwise, I will not know how to properly caption my own character for subsequent training.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.