AIセルシェーディングジェネレーター

プロンプトからセルシェーディングの画像とアニメーションを生成:フラットな色彩、シャープな影、クリーンな輪郭で、手描きアニメーションセルのような仕上がり。無料でお試しいただけます。

AIセルシェーディングの理由

セルシェーディング(トゥーンシェーディングとも呼ばれる)は、滑らかなグラデーションではなく、エッジの効いた数段階のフラットなトーンで絵を照らし、手描きアニメーションのセルロイドシートのような見た目にします。3Dではトゥーンシェーダーとアウトラインのセットアップが必要ですが、AIではプロンプトにその質感を書き込むだけで、最初から再現できます。以下のクリップは、Walter Mauroのレジデンシー作品『Duro』、Katrin Ann Orbetaの『Mila & Sasa』、そしてこのページ用に制作した2つのゲームスタイルのフレームで、それぞれ正確なプロンプト付きです。

AIでセルシェーディングアートを作るには?

3つのステップで:シェーディングルールを含むショットを記述し、セルシェーディングスタイルを選び、1つのキャンバスで全体のショートを構築します。
  1. ステップ 1

    ショットを説明します。

    シェーディングルールをアクションの前に配置:ここでは、ソフトな絵画的背景の上にセルシェーディングのキャラクター、2コマでアニメーション、no 3D renderとno CGI smoothnessを除外することで、モデルが滑らかな3Dライトではなくフラットなトーンを維持します。

    Hand-drawn girl in a white dress spinning with a long branch in a sunlit painted forest00:07
    Walter Mauro監督『Duro』 · Seedance 2.5 · リファレンスから動画へ
    Style: hand-drawn 2D animation, cel-shaded character over a soft painted gouache background, animated on twos at 12 frames per second, choppy stepped motion cadence, brushstrokes shifting and redrawn from frame to frame, line jitter, limited warm dusty palette, no 3D render, no puppet morphing, no CGI smoothness, no lens or depth-of-field, no photoreal rendering.
    
    @Image1 as character reference — DÚRÓ, exact 1-to-1 face, cornrows with golden beads, white cotton dress, golden fish pendant. @Image3 as background environment — FLORESTA, Àiyé forest, dense ancient trees, golden-hour light filtered through canopy, hard-edged dappled shapes.
    
    Camera acts as a rostrum over the artwork — slow lateral pan following her spin, one motivated move only, never handheld, never a lens move.
    
    Shot 1: Dúró grips a long wooden branch with both hands and spins in a full joyful circle. A burst of fully animated motion carries the turn — every frame hand-redrawn, dress hem flaring outward, cornrow beads swinging wide, dust kicked up from bare feet. Speed lines streak faintly behind the branch's arc. The spin lands on a single high-contrast impact frame: her body fully turned, branch raised, wide open laughing expression, held for a beat. Then follow-through settles — the dress hem keeps swaying and drops after her body has already stopped, loose braid strands keep drifting for a moment before finally going still.
    
    Sound: a light whoosh on the branch's swing, a soft giggle at the impact frame, bare feet padding dust, forest ambience underneath.
    
    CONSTRAINTS: DÚRÓ's face, cornrows and dress match @Image1 exactly 1-to-1; FLORESTA matches @Image3 throughout; no modern elements, no plastic, no wristwatch, no printed logos; avoid jitter, avoid bent limbs, avoid identity drift, avoid temporal flicker, avoid chaotic composition; no lens, bokeh or depth-of-field vocabulary; no text, no watermark; No music. Only SFX.
    意味:ドゥロという少女が森で木の枝を持って回転し笑い、手描き2Dアニメ風のセル画と柔らかいガッシュ背景で描かれています。
  2. ステップ 2

    スタイルを選択します。

    セルシェーディングスタイルを一度選べば、すべてのショットがその輪郭、フラットな色彩、影の形状を継承します。1990年代のアニメセルから大胆なグラフィックカートゥーンまで、それぞれFlickpediaで開きます。

  3. ステップ 3

    ショートフィルム全体を構築。

    各ショットをキャラクターのリファレンスと元となった絵画的背景に接続することで、同じセルシェーディングの少女が映画全体を通して登場します。これはWalter Mauroの手描きレジデンシー作品『Duro』のキャンバスの一部です。

    Walter MauroによるDuro

    ドラッグで移動、スクロールでズーム。

    作品と完全なワークフローを見る

Flickで制作されたセルシェーディング動画

ペイント調の2Dキャラクター、フラットなパステルカートゥーン、ゲームスタイルのフレーム。それぞれシャープな影のエッジを持つ数段階のフラットなトーンで照らされ、1つのプロンプトから生成されています。
Hand-drawn girl in a white dress running through tall grass toward a wooden hut with a small brown dog00:15

少女と犬が家に向かって駆ける

キャラクターは2段階のフラットなトーンでセルシェーディングされ、草と小屋は豊かな絵画調のまま。これは古典的な手描きアニメーションの特徴的な表現の分離です。
Style: hand-drawn 2D animation, richly painted cel-shaded character over a highly detailed painted background, confident linework with painterly edge variation — not vector-clean, not stiff — dramatic directional lighting with strong contrast, dense layered background depth, textured but controlled brushwork, animated on twos at 12 frames per second, every single frame independently hand-redrawn with visible line jitter and boil, choppy stepped motion cadence, no 3D render, no puppet morphing, no CGI smoothness, no motion blur, no lens or depth-of-field, no photoreal rendering.

@Image1 as character reference — DÚRÓ, exact 1-to-1 face, cornrows with golden beads fixed to scalp, white cotton dress, golden fish pendant. @Image7 as character reference — COELHO, small brown rabbit with a yellow ribbon bow at its neck, exact 1-to-1 fur color and bow. @Image6 as background environment — CASA, wooden house exterior.

Off-center framing throughout. Nothing holds fully still — soft blinks on twos, chest breathing, fur and hair drifting.

Shot 1 — WIDE, off-center. Dúró lunges forward and tumbles playfully onto the grass, catching the rabbit mid-air and rolling with it, wrapping both arms around it in a joyful hug as she comes to rest on her side, dust and loose grass kicked up around the fall, her laughter visible in her open smile.

HARD CUT to Shot 2 — POV SHOT from the rabbit's perspective, off-center, strong backlight. Dúró lies on her side in the grass looking directly toward camera, holding the rabbit close to her chest just out of frame at the bottom edge, warm morning sun behind her turning her cornrows and the loose strands around her face into a glowing golden rim light, her expression soft and content, dust motes drifting through the backlit air.

HARD CUT to Shot 3 — MEDIUM, off-center. The rabbit wriggles and slips free of her arms, hopping a short distance away and turning to look back at her. Dúró sits up fast, half-surprised, half-delighted, a startled laugh breaking across her face, one hand still reaching out where the rabbit had been.

CONSTRAINTS: three shots, two hard cuts, not one continuous take; DÚRÓ matches @Image1 exactly; the rabbit matches @Image7 exactly including the yellow ribbon; CASA visible in soft background in Shot 1; Shot 2 is a true POV from the rabbit's eye level, Dúró's hands holding it are visible but not the rabbit's own body; off-center framing only, no centered symmetry; no modern elements, no plastic, no printed logos; avoid jitter, avoid bent limbs, avoid identity drift, avoid temporal flicker, avoid chaotic composition; no lens, bokeh or depth-of-field vocabulary except the specified backlight glow in Shot 2; no text, no watermark; No music. Only SFX (soft thud of the fall, grass rustling, rabbit's small squirm, Dúró's laughter, birdsong).
duration: 15s (three shots, force explicit duration, do not leave automatic)
Walter Mauro監督『Duro』
意味:ドゥロが草の上で兎を捕まえて転がり抱きしめ、兎の視点ショットを挟み、兎が逃げる3つのシーン、手描き2Dアニメ風です。
Two hand-drawn girls holding hands and running through pink and orange sunset clouds00:30

夕焼け空の二人の少女

顔とドレスのはっきりとした影が、柔らかく絵画的な空を背景に人物を鮮明に保っています。
Style Prefix: Hand-drawn 2D animation, richly painted gouache and watercolor texture, cel-shaded characters over a highly detailed background, confident brushwork, animated on twos at 12fps, line boil aesthetic, no 3D render, no CGI smoothness, no motion blur, no depth-of-field.

Character References: DÚRÓ (@Image1) exact 1-to-1 face, cornrows, white cotton dress, golden fish pendant, normal human eyes. DUPLO/ENIKEJI (@Image2) identical face, white facial paint dots, blue patterned dress with gold sash, glowing solid bright yellow eyes.

Lighting & Environment: Radiant golden hour sunset with intense volumetric sunbeams (god rays) cutting through rosy sky matching input image 1. Strong backlit rim lighting illuminating character skin, hair braids, and dress folds. Small, delicate, natural-sized iridescent soap bubbles floating naturally matching input image 2.

Action Sequence (30 Seconds Total - 4 Shots):
• Shot 1 (Medium-Wide, 6s): Enikeji (@Image2) and Dúró (@Image1) glide gracefully side-by-side through the vibrant golden sunset haze, holding hands gently as wind ripples their clothes.

• Shot 2 (Immersive 2.5D Close Push-Through, 8s): HARD CUT to Dúró (@Image1) flying dynamically straight toward the lens, her white dress and braided hair flowing backward dramatically in 2.5D depth. Golden sunbeams rim-light her face as she speaks in Portuguese: "UAAAU, respira fundo... que lugar LINDO!" right before sweeping past the edge of the camera frame.

• Shot 3 (Close Dynamic Tracking Shot, 8s): HARD CUT to a tight tracking camera moving fluidly right beside Dúró and Enikeji as they playfully swoop and weave through the air. The camera stays very close, following their free, harmonious movement side-to-side with soft depth effect as small natural soap bubbles drift past.

• Shot 4 (Warm Slow-Motion Emotion & Facial Focus, 8s): HARD CUT to extreme slow-motion close-up focusing intensely on the joyful faces of Dúró and Enikeji smiling brightly at each other. Rich golden sunset light washes over their skin with high-contrast volumetric glow. A tiny iridescent soap bubble drifts smoothly across the immediate foreground lens, creating an intimate, magical frame.

Constraints: Exact 1-to-1 reference fidelity to @Image1, @Image2, input image 1, and input image 2; DÚRÓ's eyes fully normal; ENIKEJI's eyes glow bright yellow; small natural bubbles only; off-center framing; no text, no watermark. Audio language: Portuguese. Dialogue in Shot 2: Dúró says "UAAAU, respira fundo... que lugar LINDO!". SFX: soft wind rush, warm ambient hum.
Walter Mauro監督『Duro』
意味:手描き2Dアニメで夕焼けの空を二人の少女が手を繋いで飛び、シャボン玉が漂い、ポルトガル語で会話する物語です。
Flat 2D cartoon of a blue-haired girl and a small blue cat-suit character sharing a tart in a desert00:15

難破船でタルトを分け合う

平坦なパステルの塗り、一つの影のトーン、細いアウトラインが、テレビアニメのようなクリーンでグラフィカルなセル画の見た目を生み出しています。
Start and maintain background @Mila in Spaceship Interior 100%.
Mila's appearance @Mila con Zaino Giapponese, Senza Ombre 100% and expressions @Mila - Espressioni del Viso.
Mila's voice @My Radar Apology.
Sasa's appearance @Sasa con Zaino Giallo e Pistola Laser 100% and expressions @Sasa - Espressioni Facciali 100%. Remove guns from Sasa. Remove any glow form the characters.
Sasa's voice @Mother's Old Spaceship. 2D hand-drawn animation.
Sasa is smaller than Mila. Tart looks like this @Mila con Crostata. Do not take this image background. The tart is 90% smaller. No glow on their heads.
Preserve reference style and character identity. CAMERA: 2 shots, 1 hard cut. SHOT 1 — Medium close-up, 3/4.Mila and Sasa sit side by side on the top of the spaceship. Mila in front then Sasa beside her. Mila holds a tart sliced into 6 pieces. In the center the tart the star spice in the center.
Sasa, happy for Mila with deep warm paternal male voice: “What were the odds we'd find your mother's old spaceship?” Mila turns to him, calm, visibly happy, smiles softly: “I’ve found my star-shaped spice.” Close up - Mila takes a slice with the star spice on top of the slice and gives a bite to it and closes her eyes. The tart has 5 slices left. She closes her eyes tasting the tart. Close up of Mila with closed eyes and and tasting the tart, she opens her eyes and looks up the sky. Her expression becomes quietly hopeful: “But I have to find her among the stars.” AUDIO: Natural dialogue, no overlap. Sasa's voice is deep, warm and human, never robotic. Mila has natural voice, never robotic. No music. No sound. Hand-drawn 2D animation cozy retro-futuristic sci-fi, warm pastel palette, soft sepia linework, vintage children's book, gentle golden light, whimsical, nostalgic, cozy spaceship interior, cinematic parallax, subtle atmospheric animation.
Katrin Ann Orbeta監督『Mila & Sasa』
意味:宇宙船の上でミラとササが母の宇宙船と星型スパイスのタルトについて話し彼女が母を星々の中に探すと言う手描き2Dアニメ。
Flat 2D cartoon of a mother putting a tart into a freezer while a small child watches in a sunny kitchen00:15

冷凍庫の前の母と子

グラデーションをほとんど使わない柔らかなクリームとブルーの塗りが、家庭的なシーンを平坦で温かく、セル画のように保っています。
Use Mamma e Bambina con Crostata, Freezeras reference. One continuous animated 2D scene with a fixed camera. The mother gently places the tart inside the freezer and says: “You can keep it in the freezer for up to a month. And when you thaw it, it’ll still taste just as good.” Zoom in to the child looks at her, suddenly realizes something, and asks: “So… it’s like freezing time?” The mother smiles softly and replies: “Exactly. You can save a little moment and come back to it whenever you want.” Use natural movements, subtle expressions, and accurate lip sync. Preserve the characters, background, colors, and illustrated 2D style. No cuts, camera movement, or deformation.

Katrin Ann Orbeta監督『Mila & Sasa』
意味:イラスト風2Dアニメで母親が冷凍庫にタルトを入れ、娘が時間を凍らせるみたいと気づく温かい親子の会話場面を描きます。
赤いスカーフとゴーグルをつけたセルシェーディングの少女がホバースクーターでオレンジ色の砂漠の峡谷を走る

峡谷のゲームスタイルのライダー

人物の二つの平坦なトーン、はっきりとした影の線、リムライトが、セルシェーディングのゲームカットシーンのような見た目を生み出しています。
A frame from a 3D animated game cutscene with cel shading: a girl in a red scarf and flight goggles rides a hover scooter across a sunlit desert canyon, her model lit in two flat tones with a hard-edged shadow line instead of a smooth gradient, thick black outlines traced around every silhouette, a crisp rim light on her shoulders, and a flat painted sky and cliffs in bold oranges and teal. No text, no logos, no interface.
意味:セルシェーディング3Dゲームカットシーンで赤いスカーフの少女がホバースクーターで砂漠の峡谷を駆け抜ける場面です。
散らかったガレージで一つの作業灯の下、赤いバイクを修理するセルシェーディングのロボット整備士

夜のロボット整備士

距離に応じて太くなる硬い白いハイライトとアウトラインが、トゥーンシェーダーが3Dシーンに加えるものです。
A cel-shaded 3D render of a stocky robot mechanic repairing a motorbike in a cluttered garage at night: toon shading with exactly two flat tones per surface and a sharp shadow edge, bold ink outlines whose thickness varies with distance, specular highlights drawn as hard white shapes, a single warm work lamp, and flat colour on the walls and tools. No text, no logos.
意味:セルシェーディング3Dでずんぐりしたロボット整備士が夜のガレージでバイクを修理する場面を平面的な陰影で表現します。

セルシェーディングを手描きアニメーションのように動かすにはどうすればよいですか?

キャラクターが2コマ打ちで線のジッターを伴ってアニメーションされるように指定し、3Dレンダリングやモーションブラーを除外してください。ビデオモデルは指定しなければセルシェーディングの人物を3Dのように滑らかに動かしてしまうためです。
  • リズムを指定する:animated on twos、choppy stepped motion cadence。
  • 線を生き生きと保つ:brushstrokes shifting and redrawn from frame to frame, line jitter。
  • 3Dルックを排除:no 3D render, no CGI smoothness, no motion blur。
Hand-drawn girl in a white dress walking down a forest path through shafts of golden light00:07

木漏れ日の森を歩く

Style Prefix:
Style: hand-drawn 2D animation, richly painted cel-shaded character over a highly detailed painted background, crisp confident linework closing every shape, dramatic directional lighting with strong contrast between lit and shadowed areas, dense layered background depth with multiple sharp planes, textured but controlled brushwork, high production polish, animated on twos at 12 frames per second — each drawing held for two frames then replaced, choppy stepped motion cadence, brushstrokes shifting and redrawn from frame to frame, line jitter, limited warm dusty palette, painted light bloom, no 3D render, no puppet morphing, no CGI smoothness, no motion blur, no lens or depth-of-field, no photoreal rendering.

@Image1 as character reference — DÚRÓ, exact 1-to-1 face, cornrows with golden beads fixed to scalp, white cotton dress, golden fish pendant. @Image2 as character reference — ENIKEJI, identical face to Dúró but serene, intense silvery rim light drawn as a clean glowing shape along her silhouette. @Image3 as background environment — FLORESTA, Àiyé forest, dense ancient trees, golden-hour light filtered through canopy, painted as hard two-tone cel shadow. @Image4 as background environment — VILA, mud-and-thatch settlement, warm ochre dusk light.

Physical world (Àiyé): warm dappled light drawn as hard-edged shapes through leaves, earthy palette — greens, ochre, terracotta, gold. Spiritual world (Ìrònà/Òrun): diffuse silvery-blue light with no cast shadows, drifting golden particles drawn as small light shapes, translucent painted quality.

Camera acts as a rostrum over the artwork — slow push-in on a held face, pan across the painted background, vertical tilt down through the canopy; one motivated move per shot, never handheld, never a lens move. Off-center framing as a default — avoid centered symmetrical composition unless a shot specifically calls for it.

Nothing holds fully still — soft blinks on twos, chest breathing, loose strands drifting, foliage stirring, particles in constant motion. Cornrows render as one bundled shape at all times, using a single motion smear at peak speed instead of separating into individual strands; only the beaded tips swing free; cornrow pattern fixed to the scalp exactly as in @Image1 at rest.

Constraints:
2D hand-drawn on twos at 12fps with painterly boil, never gliding; crisp clean linework not loose sketch lines, high contrast dramatic lighting not flat even light, detailed layered background not soft blur; NO 3D render, NO puppet morphing, NO motion blur, NO liquid/melting surfaces (except where a shot explicitly names an exception, and never on Dúró's own body); NO lens, bokeh, depth-of-field or camera-body vocabulary; exact 1-to-1 reference fidelity to character and environment images; no modern elements — no plastic, no wristwatch, no rubber-sole shoes, no printed logos, no visible technology; nothing frozen or static at any point; centered symmetrical framing only when explicitly called for; avoid jitter, avoid bent limbs, avoid identity drift, avoid temporal flicker, avoid chaotic composition; No music. Only SFX.
Walter Mauro監督『Duro』
意味:スタイルプレフィックスとして手描き2Dアニメの制作ルールと撮影技法とキャラクター環境参照と世界観の照明設定を定義しています
Close up of a hand-drawn girl's face with braids, smiling, in warm painted light00:22

2コマ撮りのクローズアップ

Style: hand-drawn 2D animation, richly painted cel-shaded character over a highly detailed painted background, confident linework with painterly edge variation — not vector-clean, not stiff — dramatic directional lighting with strong contrast, dense layered background depth, textured but controlled brushwork, animated on twos at 12 frames per second, every single frame independently hand-redrawn with visible line jitter and boil, choppy stepped motion cadence, no 3D render, no puppet morphing, no CGI smoothness, no motion blur, no lens or depth-of-field, no photoreal rendering.

DIRECTOR'S NOTE — lens optics exception for Shot 1 only: clear sharp focus on Dúró's face with a tiny golden reflection of the rabbit visible in her eyes. The barrier is invisible until struck or pressed.

@Image1 as character reference — DÚRÓ, exact 1-to-1 face, cornrows with golden beads, white cotton dress, golden fish pendant. @Image7 as character reference — COELHO, small brown rabbit with a golden beaded cord. @Image3 as background environment — dense forest path with strong sunlight beams hitting the ground.

Off-center framing throughout. Nothing holds fully still.

Shot 1 — CLOSE-UP, off-center (5s). Dúró's face lights up with a warm, hopeful smile. In a detailed close-up on her eyes, a tiny reflection of the small brown rabbit sitting in the sunlit path is clearly visible in her irises.

HARD CUT to Shot 2 — MEDIUM-WIDE, off-center (6s). Dúró suddenly breaks into a run toward the rabbit. Mid-stride, she collides abruptly with an unseen wall in mid-air — the invisible barrier ripples violently with iridescent light upon impact. Ahead on the sunlit path, the rabbit sits perfectly calm, watching her.

HARD CUT to Shot 3 — LOW ANGLE MEDIUM, off-center (5s). Knocked back by the shock, Dúró falls onto the dirt path, sitting on the ground. She touches the earth and stares upward into the blank air in front of her, completely bewildered and confused.

HARD CUT to Shot 4 — MEDIUM, off-center (6s). Dúró gets up and presses both open palms firmly against the empty air. The barrier activates under her weight, bulging and shimmering like a thick iridescent gel wall that resists her push and holds her back.

CONSTRAINTS: four shots, three hard cuts, not one continuous take; DÚRÓ matches @Image1 exactly; the rabbit matches @Image7 exactly; forest matches @Image3; Dúró remains in the forest; off-center framing only; no modern elements; avoid jitter, avoid bent limbs, avoid identity drift; no text, no watermark; No music. Only SFX (gasp of joy, sudden thud/impact, dirt rustle from fall, heavy shimmering hum on push).

duration: 22s (four shots: 5s, 6s, 5s, 6s; force explicit duration, do not leave automatic)
Walter Mauro監督『Duro』
意味:三角髪の少女が森で走り出すが透明な壁にぶつかり倒れ両手で押すと虹色の壁が現れる手描きアニメ風映像が作られる

AIイラストにセルシェーディングを与える言葉は?

スタイルだけでなくシェーディングも指定:2〜3段階のフラットトーン、硬いエッジの影、クリーンな輪郭線。「anime」という言葉だけでは、柔らかなデジタルグラデーションになりがちです。
  • シェーディング:cel-shaded、two flat tones of shadow、hard-edged shadow line instead of a smooth gradient。
  • 線:crisp outlines closing every shape、または軽やかなTVアニメ風のthin outlines。
  • 光:rim light、specular highlights drawn as hard white shapes。
  • キャラクターと背景を分離:cel-shaded character over a highly detailed painted background。

AIセルシェーディング生成に関するFAQ

セルシェーディングとは?

セルシェーディング、またはトゥーンシェーディングは、滑らかなグラデーションの代わりに数段階のフラットトーンと硬いエッジの影で画像をレンダリングする手法で、多くの場合暗い輪郭線を伴い、3Dシーンやデジタルアートを手描きアニメーションセルのように見せます。

なぜセルシェーディングと呼ばれるのか?

この名前は、伝統的な2Dアニメーションでアニメーターがキャラクターを描いた透明なセルロイドシート「セル」に由来します。セルシェーディングは、その塗られたシートの平坦で硬いエッジの色合いを模倣したものです。

AIでセルシェーディングアートを作成できますか?

はい。Flickはテキストの説明からセルシェーディングの画像や映像を生成します。フラットな色調、くっきりとした影、クリーンな輪郭を指定し、セルシェーディングスタイルを選択すれば、すべてのショットで同じ見た目を維持できます。

セルシェーディングとアニメの違いは何ですか?

アニメは日本発祥のアニメーションスタイルで、セルシェーディングは照明技術です。伝統的なアニメの多くはセルシェーディングですが、セルシェーディングは西洋のカートゥーンや3Dビデオゲームでも使用されています。

セルシェーディングアニメーションを作成するにはどうすればよいですか?

セルシェーディングの静止画またはキャラクターリファレンスを生成し、Flickの映像モデルでアニメーション化します。その際、キャラクターを2コマ打ちでアニメーション化し、3Dレンダリングやモーションブラーを除外することで、フラットなシェーディングを維持できます。

セルシェーディングジェネレーターは無料ですか?

はい、無料で始められます。Freeプランではクレジットカード不要で毎週100クレジットが付与され、GPT Image 2.6が利用できます。Seedance 2.5などの映像モデルにはStandardプラン(月額20ドル)が必要です。

Flickのセルシェーディングスタイルをもっと見る

Flickでセルシェーディングスタイルを試す

無料でお試しいただけます。クレジットカードは不要です。

このページについてAIに質問:ChatGPTPerplexityClaude