日本語での説明は下にあります。Trigger words and prompts notes are down below.This is a concept LoRA for enhancing the doyagao tag with a simple prompt. Doyagao can be considered as smile / smug / smirk with v-shaped eyebrows. While the original &gt;:) tag in the base model can generate doyagao images, I feel these are more offensive than I expected. That's why I made this model.It supports all four variants; open eyes + closed mouth, open eyes + open mouth, closed eyes + closed mouth and closed eyes + closed mouth.All training images are female, so it would be challenging for making doyagao male images.Unfortunately, because of limited sources, generating the image from side, looking back from behind or looking to the side would be weak.This LoRA is based on <a target="_blank" rel="ugc" href="https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0">Illustrious-XL 0.1</a> and used <a target="_blank" rel="ugc" href="https://civarchive.com/models/823108?modelVersionId=1564750">AnyIllustrious-XL(for LoRA training)</a>. I recommend using <a target="_blank" rel="ugc" href="https://civarchive.com/models/820208?modelVersionId=1136462">Obsession (Illustrious-XL) v3.1 </a>or <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-illustrious-sdxl?modelVersionId=1761560">WAI-illustrious-SDXL</a>(technically the newer of these are <a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL</a> based models)<h1 id="trigger-words-and-prompts">Trigger Words and Prompts</h1><h2 id="trigger-words">Trigger Words</h2><ul><li>doyagao</li><li>doyagao, open mouth</li><li>doyagao,closed eyes</li><li>doyagao, closed eyes, open mouth</li></ul>When the model failed to generate closed smile: add :dWhen the model failed to generate open smile: add c:You may add smug or smirk to enhance these attitudes.<h2 id="tag-examples">Tag Examples</h2><h3 id="effects">Effects</h3><ul><li>puff of air</li><li>sparkle</li><li>emphasis lines</li></ul><h3 id="pose">Pose</h3><ul><li>crossed arms</li><li>hands on own hips</li><li>hand on own chest</li><li>v over eye</li><li>v over head</li></ul><h3></h3><hr />トリガーワードとプロンプトの注意点は下にあります。これはドヤ顔のLoRAです。doyagao タグを補助してなるべくシンプルなプロンプトにすることを目標としています。&gt;:) タグでドヤ顔画像を作成できますが、生成された画像がやや攻撃的な表情に見えたのがこのモデルの作成理由です。このモデルは4つすべてのバリエーションをサポートしてます。つまり、開眼 + 閉口、開眼 + 開口、閉眼 + 閉口と閉眼 + 開口です。教師画像がすべて女性なので、男性キャラのドヤ顔をつくるのはすこし難しいかもしれません。残念ながら教師画像が限定されているため、横から・後ろから振り返る・よそを見るの画像生成は弱いかもしれません。このLoRAのベースモデルは <a target="_blank" rel="ugc" href="https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0">Illustrious-XL 0.1</a> です。 <a target="_blank" rel="ugc" href="https://civarchive.com/models/823108?modelVersionId=1564750">AnyIllustrious-XL(for LoRA training)</a>を使って学習させました。<a target="_blank" rel="ugc" href="https://civarchive.com/models/820208?modelVersionId=1136462">Obsession (Illustrious-XL) v3.1</a>か<a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-illustrious-sdxl?modelVersionId=1761560">WAI-illustrious-SDXL</a>の使用を推奨。(どちらも新しいバージョンは<a target="_blank" rel="ugc" href="https://civarchive.com/models/833294?modelVersionId=1116447">NoobAI-XL</a>ベースのモデル)<h1>トリガーワードとプロンプト</h1><h2>トリガーワード</h2><ul><li>doyagao</li><li>doyagao, open mouth</li><li>doyagao,closed eyes</li><li>doyagao, closed eyes, open mouth</li></ul>閉じた笑顔の生成に失敗するときは :d を追加開いた笑顔の生成に失敗するこきは c: を追加smug や smirk を追加して、よりドヤ感を出すこともできます<h2>タグ例</h2><h3>エフェクト</h3><ul><li>puff of air</li><li>sparkle</li><li>emphasis lines</li></ul><h3>ポーズ</h3><ul><li>crossed arms</li><li>hands on own hips</li><li>hand on own chest</li><li>v over eye</li><li>v over head</li></ul>

Files with SHA256: a05b382ce55ee93effd6b89ae12c08dd9df6b014cad087a91e6c12657b52c22d