UPDATE : I built a quick ADetailer model for phones. It works well so far on images generate with this Lora and will probably pick up some other phones too. It's not perfect but it's usable.
If there's more interest I'll keep training the ADetailer.
Pick it up in the extra files section (it's the small Lora, I cant seem to rename it)
Drop it in /models/adetailer and restart auto1111
***********
Finally made something that has been missing for a long time. Proper phones in selfies!
I focused on building a very carefully curated dataset of one specific iPhone type.
The Lora does a pretty good job of adding it.
keywords are : "iphone mirror selfie" and helped along by "<subject> holding iphone"
Have fun! Please post your creations here :)
Description
FAQ
Comments (17)
it will sometimes generate a black iphone or black iphone cover... I didnt train it on anything but the base white/silver phone with clear case. I have no idea where the other phone comes from. It's kinda cool though...
Hi!
could you make a version for 1.5 or provide your dataset so i could make it for 1.5? :)
thanks in advance :)
@bicpen11 sdxl is not compatible with 1.5, those are two different lines of models
If you can combine it with "PE Holding Sign [Concept]".You're ready to Fishcat :)
Is both the safeTensor and pt version needed (lora and embedding)?
Or is one just more condensed than the other?
Or something else (lite vs full)?
The pt version is a adetailer model. I just could not figure out how to rename them on the file list... Will try again later
1.5 pweease
this works ridiculously well
PLZ do a 1.5 one! <3
Are both files needed? And where to put them? In Lora folder right since one is adetailer, I have an adetailer folder as well.
both arent needed. The large one is the LORA which goes in LORA folder. The small .pt file goes in your adetailer folder if you want to use that.
yup m2 any soliton?
Then it's probably not compatible with your checkpoint model.
Hello, do 2 trigger words also need to be used, and what should be their weight?
I've used Lora quite a bit, but the main problem is that the details on the phones are really bad. From a distance, there's nothing wrong, but when you get up close, you see some pretty serious errors. If this is fixed, Lora will be a very good solution.
its very old, small probability it will get updated
Details
Files
iphone_mirror_selfie_v01b.safetensors
Mirrors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
iphone_mirror_selfie_v01b.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.