すべて · SNS投稿 · 季節・行事 · 共感ネタ
259わたしの写真でハロウィン仮装3Dキャラ — 魔女・ヴァンパイア・かぼちゃ王子に変身
写真がハロウィンの仮装をしたかわいい3Dキャラクターに。仮装を選ぶだけで、かぼちゃのお菓子かごとおばけの友だちまで一枚に入ります。
ChatGPT 無料 OK
項目を入力
- 1項目を入力します
- 2プロンプトをコピーします
- 3GPTのチャットに貼り付けます
- 4そのまま送ると、サムネイルの例のような画像になります
- 5自分の写真一枚をプロンプトと一緒に入れると、その写真が反映されます 任意
プロンプト全文を見る 8,869文字
Generate this image now. Draw it directly - do not describe it, explain it, or ask questions. 出力サイズは1080×1350ピクセル、4:5の縦型Instagram投稿にしてください。 PHOTO RULE (this overrides any other line in this prompt, before or after it, that calls the photo required or mandatory or says to ask for it): if a photo is attached, use it exactly as the instructions below describe. If no photo is attached, never ask for one and never stop — invent a fitting fictional subject (person, pet, food, product or place, whatever this card is about) in the spirit of the card's example image, and finish the image. [Resolve the costume first] Create ONE finished cute, original 3D animated Halloween character portrait. The uploaded person becomes the recognizable protagonist, shown full-length or from the knees upward in a complete chosen costume. The size on line two is authoritative; default to portrait 4:5. Deliver one polished image, not a before/after, character turntable, several costume options, phone mockup or printed greeting card lying on a table. COSTUME = "【仮装】". This required visual selection defines ALL wardrobe, headwear, shoes and costume accessories before any scene is built. The person's original everyday clothing is replaced by this costume; no later sentence may restore a different costume. EXPRESSION = "【表情】". If absent or photo-unchanged, retain the reference expression. A playful wink closes one eye with a mischievous smile; pretending to growl is a funny puffed-cheek theatrical face, never aggression; surprised means raised brows and a small rounded mouth. HAND_PROP = "【持っているもの】". If absent, use one pumpkin candy basket. A broom, lantern or lollipop selection replaces the held basket completely; never force both selected and default handheld objects into the hands. BACKGROUND = "【背景】". If absent, use purple night sky with pumpkin lanterns. The selected setting controls the whole environment, not an added miniature panel. GREETING = "【あいさつのひとこと】". If absent, print no words. If automatic, choose one short friendly Halloween greeting. Otherwise print this input verbatim once, without translating or adding a second slogan. [Fallbacks for absent input lines] When an optional named input line is absent, resolve EXPRESSION to the reference expression (or the fictional subject's natural friendly expression), HAND_PROP to one pumpkin candy basket and BACKGROUND to purple night sky with pumpkin lanterns. GREETING is absent unless its own line is supplied. Active lines above always take precedence over this fallback map. [Input priority] Resolve the named inputs BEFORE planning the image. An optional input exists only when its own input line is supplied; blank values and "選ばない" mean absent. "おまかせ" are automatic-choice MENU COMMANDS, never visible copy: choose a fitting concrete result yourself. Never print a selector, field label, these commands, or placeholder braces. Defaults apply only to absent inputs. A supplied choice replaces its corresponding default everywhere; subsequent scene descriptions refer to the resolved named value, never to a competing fixed default. Use the following costume interpretations only for the resolved COSTUME. Do not draw the menu, name of the costume or a list of options. A costume may have its own worn accessories, but the only held object is HAND_PROP. [Costume construction] Cute witch: an original soft pointed hat with a bent tip, a plum tailored coat or tunic, a short scalloped cape and rounded ankle boots; no broom unless HAND_PROP selects it. Vampire count: an original high-collar cape with deep wine lining, a formal dark vest, a pale shirt and fitted trousers or a simple formal skirt consistent with the person's presentation; no pointed witch hat. Pumpkin prince/princess: a small leaf-shaped crown, a rounded pumpkin-orange royal tunic or dress with ribbed fabric panels, a green collar and soft boots; select an appropriate royal presentation without changing the person's identity, and never reuse witch clothing. Little ghost: a soft rounded white ghost poncho with the person's entire face visible, gentle hem folds and simple shoes; no covered eyes or replaced face. Black cat: a charcoal cat-ear hood or headband, soft textured fitted costume, rounded paw-like gloves and an attached curved costume tail; preserve a human face rather than replacing it with an animal. Mummy: clean ivory costume bandages over a modest base garment, wrapped limbs and a loose end at one shoulder, with the face and eyes uncovered; no wounds or decay. Ghost pirate: an original weathered tricorne, a sea-green coat with a translucent ghostly cape edge, a plain sash and soft boots; no weapon, logo or existing pirate-character outfit. In every selection retain the photographed subject's age and use comfortable, nonsexualized clothing. [Identity source] An attached photograph supplies facial identity, age, gender presentation, skin tone, eye shape and spacing, nose, mouth, facial outline, hairstyle, glasses, freckles, facial hair and any disability aids. Preserve these recognizable features through stylization. Round the shapes gently and slightly enlarge expressive eyes while keeping their relative placement; do not replace the subject with a generic baby, doll, famous performer or stock animation face. Preserve eyewear even under a hat, and let the real hairstyle emerge naturally. If a photo contains several people, use the most prominent clearly visible person, chosen by central placement and visible facial detail, without adding the others. Without a photo, follow PHOTO RULE and invent a fitting fictional subject immediately. Identity comes from the photo; wardrobe comes only from COSTUME. [Composition and environment] The protagonist occupies around 65-75% of the frame height with a readable costume silhouette, comfortably separated hands and room around the hat. A full-body composition includes complete feet; knee-up framing includes the costume's distinguishing lower-body details without awkward joint crops. Place one small friendly original ghost companion floating beside the protagonist's lower shoulder or hip, at less than one fifth of the protagonist's height. It has a simple expressive face and rounded cloth-like body, no accessory or speech bubble. HAND_PROP is held naturally in one hand with believable fingers, a working handle or grip and appropriate scale. If a pumpkin basket is selected, show sculpted ribs and a few plain unbranded sweets inside, with no package lettering. For a purple-sky setting, make a gently layered violet night, a few softly blurred orange pumpkin lanterns and low curling ground mist. A spooky forest instead uses friendly bent trees, cool moon shafts, a soft leafy floor and gentle haze. A candy-filled room instead uses a whimsical interior with shelves of loose, unbranded sweets and soft warm bounce light, no outdoor sky. Solid orange means a clean orange studio backdrop with a subtle ground shadow, not an outdoor scene imposed over it. The small ghost remains in every setting. Background objects support the protagonist and never block the face, selected costume or greeting. [3D art direction and lettering] Use polished feature-animation-quality rendering with original character design, soft subsurface skin shading, individually shaped hair clumps, tactile fabric seams, softly worn hat felt, warm bounce light and coherent contact shadows. The result is a rendered character, not a photoreal person or a flat vector mascot. Keep the face the sharpest area, followed by costume textures and the held object, with tasteful environmental depth. Render GREETING only if supplied, in a short softly rounded lettering line around 76-83% height on a calm background area; keep top 12% and bottom 15% free of essential lettering. Use high contrast, exact capitalization and punctuation, and no text panel covering the costume. Never render EXPRESSION, BACKGROUND or HAND_PROP as labels. [Final consistency] Check that the resolved costume has the correct hat, cape, garment and footwear and that no accessories from another menu choice remain. Keep the reference identity, age and original skin tone; use natural hand anatomy and no extra fingers or duplicated limbs. All ghosts, garments and props are newly designed, without imitating a real film, famous monster, animation studio character, franchise costume or brand. No blood, wounds, injury, frightening teeth, weapons, candy logos, watermarks or extra captions. Return the single complete Halloween 3D illustration now. [Final output format] 出力サイズは1080×1350ピクセル、4:5の縦型Instagram投稿にしてください。 This selected output format takes priority over any fixed canvas size or aspect ratio elsewhere in this template. Adapt the composition, text zones and margins to the selected canvas without stretching or cropping essential content. Device safe areas apply only when a device is selected. Never print these format instructions on the image.
結果はどうでしたか?
使い方・よくある質問
こんなときに
ハロウィンの仮装はしたいけれど、衣装を買ったり出かけたりできないときに使ってください。写真が、魔女・ヴァンパイア伯爵・かぼちゃ王子などの仮装をしたかわいい3Dアニメーションキャラクターになり、かぼちゃのお菓子かごと小さなおばけの友だちが隣に付きます。10月のプロフィール写真、インスタのストーリー、グループチャットのハロウィンのあいさつにぴったりです。
項目の入れ方
仮装を一つ選ぶだけです。ほかを空欄にすると、写真の表情・かぼちゃのお菓子かご・紫の夜空とかぼちゃランタンの背景になります。あいさつは絵の下に小さく入ります。短い英語がいちばん正確に出ます。
写真の入れ方
① 顔がよく見える写真をアップします — まだ送りません
② 同じ欄にコピーしたプロンプトを貼り付けます
③ 両方入ったのを確認して送ります
確認すること
- 顔が自分に似ていますか?髪型・メガネ・ほくろが反映されているか
- 選んだ仮装になっていますか?マント・帽子・小物がその仮装らしいか
- あいさつの文字は正しいですか?
直したいとき
同じ会話で一つずつ頼んでください。「顔を写真にもっと似せて」「帽子をもっと大きく」「背景だけ不気味な森に」のように。
よくある質問
- 写真なしでもできますか? できます。新しい架空のキャラクターで描きます。
- 子どもの写真でもいいですか? できますが、子どもの写真は公開SNSより家族のグループチャットで使うことをおすすめします。








