Trained with around 150 LUTs on a small dataset, will try to expand it for the next iteration.
Kind of captured its meaning, makes photos look like you applied a filter to them and can pick up on adjectives as well.
Base model used for training: ChimeraMix V5 - so should be versatile, leaning towards realism, high native res. support (~960x960)
Description
~200-300 different LUTs applied to 4 different photos + some ground truth data
will try to increase variety for the next one
FAQ
Comments (9)
Looks great. It is possible to create Lora of Luts?
the current one isn't good to share, it needed to be merged in to work properly, but I'll try to make a LORA too from the next one
@wier yes please, we need lora
Thank you very much for answering. Is it possible to create a lut similar to the color of the Midjourney images, does it seem somewhat cinematic?
@marcoescobedo good idea, i'll try and see if it works
@marcoescobedo @nishu i finished it: https://civitai.com/models/101036
added some midjourney images, tagged as Midjourney LUT, not sure how well it works, but it certainly changes some stuff
@wier thanks for your contributions
looks over saturated, but will respond their after trials :)
thanks again!
Thanks, I'll try with some pictures.
❤️❤️❤️❤️❤️❤️❤️❤️❤️So beautiful 😍😍😍😍😍😍😍
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.