Creator: ayasesuki
The English description is translated below the Japanese one at DeepL.
髪をかきあげて顔の横に来ている髪を耳にかけて後ろに流す仕草のLoRAです。会話中、食事中、飲み物を飲む、本を読むなど様々なシチュエーションで活用できます。
トリガーワードは tucking hair です。
重みは0.7前後推奨。重みを上げると破綻が増えやすくなりますが両手率が下がり、下げると両手で弄りだす確率が上がります。
無くても動作しますが hand up, hair tucking, hand in hair, adjusting hair も推奨、LoRA無しだとぼんやりと髪を触るくらいのプロンプトですが耳を出す動作へと矯正します。
Animagine版はAnimagine XL 3.1をベースとして学習していますが3.0でも動作は確認しています。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。
Pony版はPony Diffusion V6 XLをベースとして学習していますので派生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
It is a LoRA gesture in which the hair is brushed back and the hair coming to the side of the face is swept back over the ears. It can be used in a variety of situations such as during conversation, eating, drinking, reading a book, etc.
trigger word : tucking hair
A weight of around 0.7 is recommended. Raising the weight will increase the number of failures but decrease the two-handed rate, and lowering the weight will increase the probability of two-handed tucking.
Hand up, hair tucking, hand in hair, adjusting hair are also recommended.
The Animagine version is based on Animagine XL 3.1, but we have confirmed that it works with 3.0 as well. Other XL models may not function properly.
Pony version is trained based on Pony Diffusion V6 XL, so there is a possibility that it does not work properly with derived models.