No matter how hard I tried, almost every model wants to make the subject look directly at you. I got 👁️👁️ when I wanted 🙄

I used realistic and semi-realistic images to create the V1 dataset. I focused on head and eye positions.

Strength = 0.8-1.1

Anime = 0.6 (if it turns your image real)

Trigger words

• Eyeroll

• Eyes_rolled_back

• eyeroll Rolled_up

• eyeroll to_the _side

• corner_of_her_eye (hit or miss)

• eyeroll tilts the head and looks at you from the corner of her eye.

I tried using an eye detailer, with the LoRa inline and weighted conditioning. It still did not work, so skip it.

Instagram: https://www.instagram.com/synth.studio.models/

Buy me a☕ https://ko-fi.com/lonecatone

This represents many of hours of work. If you enjoy it, please 👍like, 💬 comment , and feel free to ⚡tip 😉