camouflage

Recommended settings:

Make sure you are aware on the usage instructions of LORA

Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)

CFG Scale : 5 to 10

LORA weight for txt2img: reccomended 0.2-0.7.

Keys: wearing camouflage_uniform, wearing cap, holding a rifle, military camp, in a forest

This model i think it's a masterpiece, i really like it!!

I advice to use 0.3-0.5 as model weight.

About the main keyword "wearing camouflage_uniform" it might not be necessary, but i used it in all prompts, sometime giving to it a bigger weight.

"wearing cap" really doesn't need a weight, just use it as you want whenever you want the woman with a cap.

"holding rifle" might need an higher weight to work, i used 1.5 in my prompts. And anyway when you use this keyword be prepared to generate a lot of images before finding one that makes sense.

"military camp" and "in a forest" are just to get an image in the proper environment.

Also, as you can see from my images i have generated these uniforms in different colors, in this case, since all training images are black and green you might need to bet in SD creativity for the result and my advice is to give a very high weight to the color, like "(wearing camouflage_uniform:1.3), wearing cap, (holding a rifle:1.5), (red uniform:1.5)"

It works also with anime, i used Chillout for realistic one and Yesmix for anime ones.

I'm uploading epoch 4 out of 10, the one that gived best results with images with holding a rifle keyword (hardest image to generate). If you want to try and experience all of them come in my Patreon.