thumbnail1 / 8

typescript経由

いいね 23
eru

eru

277 Followers·1647 Posts

言語モデルを活用しマルチモーダルに意味空間を拡張した多様体に基づいた画像生成に興味があります(キリッ) また、画像を用いることにより、知恵を利用したより面白い知識を選択したより意外性のある面白い小説を生成できるかどうか、にも興味があります

2024/06/16 04:46
model name used:DALL-E3
target age:G
style:Illustration

const imageDescription = { mainSubject: { type: \"character\", features: { hairColor: \"silver\", hairStyle: \"long with twin-tails\", eyeColor: \"hazel\", clothing: { outfit: \"black long-sleeve top, red skirt with suspenders\", accessories: [\"hair ribbons\"] }, expression: \"calm and relaxed\", pose: \"lying face down on a plaid blanket, viewed from the feet side\", additionalDetails: [\"barefoot\"] } }, setting: { type: \"outdoor\", features: { blanket: \"plaid pattern\", additionalElements: [\"autumn leaves scattered around\"] } }, atmosphere: { type: \"serene\", emotions: [\"tranquility\", \"comfort\"] } };

Comments