Description
FAQ
Comments (14)
This model is INCREDIBLE... except that I am losing the head/face in like 95% of the images. even with all of this: "arms raised high. looking at viewer. cowboy shot. head shot. upper body view" i cannot get a head most of the time. Any suggestions on how to reliably get a face?
Hello. Glad to hear your comment. The easiest way is to use ControlNet. If there is no such desire, use a force value of no more than 0.7. At least, in this case, my face is always in the frame.
Use controlnet, with a pose that specifically has the face/torso. That's the only way you can guarantee it'll generate the shot you want, prompts can only go so far for that
Amazing, the results are incredible. Can you create the super short version of this ? Like the dress bottom line is incredibly short that the panties peeking through short ? Thanks for this Lora though
Hi, thanks for your comment. Ok, I'll try if I find the source)
You're doing a fantastic job with these female clothes. Keep up the amazing work!!!
Thank you very much for your positive comment!
@nextgenius I'm glad you've found a sweet spot with these LoRAs!. Have you considered doing other "tight" clothing for bottoms, like ultra-tight skinny jeans (or "jeggings"). That would be SUPER hot. You could also do some ultra tight "bodycon" skirts (or even pencil skirts). There's still so much potential :D
@BillyBobb Maybe. It all depends on whether I find a good source).
@nextgenius Fair enough. I look forward to seeing what you come up with next!!!
Great lora. Really enjoying playing around with it.
I experienced that the color of the dress is limited to brown/greyish and I have to introduce high weighted color related prompt words which has effects on the whole image.
Any chance you could improve that for a next version?
Or any tips on how to better control the dress color?
Other than setting the weight of a specific color, nothing yet. Perhaps later I will improve this feature in the next version.
really aggressively alters the style... I had prompts focusing on face but when I added the lora at .7 weight it cut out the face entirely lol
Yes, because all the models in the dataset are faceless. Otherwise, the face from the dataset will mutate the face of your own model.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.