Type | |
Stats | 2,093 79,921 8.9k |
Reviews | (374) |
Published | Feb 6, 2024 |
Base Model | |
Usage Tips | Strength: 0.9 |
Trigger Words | selfie, user interface, fake screenshot, viewfinder, phone screen, taking picture, |
Hash | AutoV2 4C672AE26E |
The English description is translated below the Japanese one at DeepL.
スマホのインカメラで自撮りしている画面のLoRAです。
プロンプトだけでも出ることは出るのですがUIが微妙なのでiPhoneを学習させました。
トリガーワードは selfie, user interface, fake screenshot, viewfinder, phone screen, taking picture, です。
やけに多いですが削ると破綻する確率が増えます。
重みは0.6~0.9くらいをお好みで。
Animagine XL 3.0をベースとして学習しているのでそれ以外のXLモデルでは正常に機能しない可能性があります。
以下翻訳
This is a LoRA of a screen taking a selfie with the phone's in-camera.
I learned iPhone because the UI is subtle, although it can appear with just a prompt.
trigger words : selfie, user interface, fake screenshot, viewfinder, phone screen, taking picture,
The number of trigger words is very large, but if you remove them, the probability of failure will increase.
The weight should be 0.6 to 0.9 as you like.
The training is based on Animagine XL 3.0, so other XL models may not function properly.