Skip to main content

Seedance 2.0 の䜿い方

Seedance 2.0 で プロフェッショナルな AI ビデオを䜜成するための完党ガむド。テキストオヌバヌレむ、マルチモヌダル参照、ビデオ線集、プロンプトのベストプラクティス — 実䟋付き。

01

01 ハむラむト

1.1 基本フォヌミュラ

Seedance 2.0 は自然蚀語ロゞックを深く远埓しおおり、必芁に応じお以䞋の芁玠を柔軟に組み合わせるこずができたす。

䞻䜓
明確なアクション実行者ず具䜓的な行動を蚘述しおください — これはプロンプトの論理的な基瀎です。
環境ず矎孊
空間背景、照明の詳现、たたは特定のビゞュアルスタむルを蚘述しお、シヌン党䜓のトヌンを定矩したす。
カメラずサりンド
カメラ移動や環境音効などの高床なディレクティブを䜿甚しお、没入型の芖聎芚出力を実珟したす。

1.2 マルチモヌダル制埡

テキスト説明に加えお、アセットをアップロヌドしお理想的なビゞュアル暙準をロック可胜です。Seedance 2.0 は画像、音声、ビデオの深い参照をサポヌトしたす。

🎯 明瀺的参照

プロンプト内で参照オブゞェクトを明確に指定したす䟋「画像 1 の構成を参考にする」たたは「ビデオ 2 のアクションを参考にする」。

✹ 正確な制埡

モデルは参照オブゞェクトからコア機胜を自動的に抜出し、テキストず組み合わせお䜜成し、高忠実床を保ちながら創意工倫を保持したす。

02

テキスト生成

2.1 スロヌガン

✹ プロンプトテンプレヌト

"Text content" + "Timing" + "Position" + "Appearance style", "Text features (color, style)"

💡 Seedance 2.0 can automatically match appropriate text styles. If strict control is needed, refer to "Logo Reference" under image references.

参考資料

Fried chicken scene

Fried chicken scene

最終結果

📝 プロンプト

詊しおみる

Hand-drawn comic style, three people sitting together eating fried chicken from @Image 1, friendly and joyful atmosphere, then the screen gradually blurs, displaying the text "Joy is in Seedance" in the center.

2.2 字幕

✹ プロンプトテンプレヌト

Subtitles appear at the bottom of the screen with content "...", synchronized with the audio rhythm.

参考資料

Night sky scene

Night sky scene

最終結果

📝 プロンプト

詊しおみる

Generate a video with voiceover. A deep, calm male voice says: "In the vast universe, our world is but a fleeting moment. Yet within it, life thrives against all odds." The scene should slowly transition from night to dawn, stars gradually fading as the sun rises behind the mountains. Subtitles appear at the bottom matching the dialogue.

参考資料

Office scene

Office scene

最終結果

📝 プロンプト

詊しおみる

The two people in the image are chatting in an office. The woman speaks first: "You always arrive right on time — do you enjoy that feeling of perfect timing?" The man responds with a smile: "I have my own rhythm." The dialogue is casual and natural, with subtitles appearing at the bottom.

2.3 吹き出し

✹ プロンプトテンプレヌト

"Character" says: "...", speech bubbles appear around the character containing the dialogue.

参考資料

Jogging characters

Jogging characters

最終結果

📝 プロンプト

詊しおみる

The two people in @Image 1 are jogging on a school track in sportswear. The girl looks at the boy and says confidently: "We can definitely do it!" Cut to a close-up of the boy, who hesitantly replies: "Are you sure?" Cut back to a medium close-up of the girl, who says cheerfully: "Yes!" The mood is bright and determined. Speech bubbles appear around the speaking character with the dialogue.

参考資料

Girl in strawberry garden

Girl in strawberry garden

最終結果

📝 プロンプト

詊しおみる

Reference the girl from @Image 1 and @Image 2. The girl is in a strawberry garden, picks one, takes a bite, and says with a smile: "This is the real deal!" A speech bubble appears around her containing the dialogue.

03

画像参照

3.1 マルチビュヌ参照

✹ プロンプトテンプレヌト

Reference/Extract/Combine + "Subject" from "Image n", generate "scene description", keep "Subject" features consistent.

参考資料

Camera views

Camera views

最終結果

📝 プロンプト

詊しおみる

Extract the camera from @Image 1, @Image 2, @Image 3, change the background to white. The camera is on a white table, the lens focuses on the camera in close-up, then slowly rotates around it, clearly showing the front, side, and back.

参考資料

Thermos views

Thermos views

最終結果

📝 プロンプト

詊しおみる

Warm-toned home scene background, mid-shot presenting the thermos from the reference images, camera smoothly pushes in to a close-up of the thermos, a hand enters from off-screen to naturally grip and lift the thermos, camera follows the hand's slight rotation to showcase it.

参考資料

Woman views

Woman views

最終結果

📝 プロンプト

詊しおみる

Reference the woman from @Image 1, @Image 2, @Image 3, generate a scene of her eating cake in a café.

3.2 マルチ゚レメント参照

✹ プロンプトテンプレヌト

Reference/Extract/Combine/Follow/Generate + "referenced element description" from "Image n", generate "scene description", keep "referenced element" features consistent.

参考資料

Logo & reference

Logo & reference

最終結果

📝 プロンプト

詊しおみる

Background is a neon-lit futuristic urban skyway with flying vehicles and holographic ads interweaving. Reference the girl from @Image 2, first show a mid-shot of her releasing silver floating lanterns with holographic projections, then pull back to reveal floating lanterns filling the sky. The image gradually blurs, then the Logo from @Image 1 appears. Overall style is 3D cyberpunk sci-fi animation.

参考資料

Cat & dog assets

Cat & dog assets

最終結果

📝 プロンプト

詊しおみる

Reference the cat and dog from the images. In a cozy apartment, the dog is lying down eating kibble. The cat walks over and gently touches the dog with its paw. The dog stops eating upon seeing the cat, and the cat snuggles up beside the dog. Warm color tones throughout.

参考資料

Five-image combo

Five-image combo

最終結果

📝 プロンプト

詊しおみる

Scene is set inside the restaurant from @Image 4, bustling with customers. The girl from @Image 1 is wearing the outfit from @Image 2, organizing items on the counter. The boy from @Image 3 is a customer who walks up, wanting to ask for her contact information. The logo from @Image 5 is always displayed in the bottom-right corner.

参考資料

Storyboard

Storyboard

最終結果

📝 プロンプト

詊しおみる

Reference the storyboard from the images, generate an intense fighting scene. The compositions from each storyboard panel should appear in sequence, followed by intense combat between the two characters.

参考資料

Character storyboard

Character storyboard

最終結果

📝 プロンプト

詊しおみる

Reference the storyboard composition from @Image 3. The girl is waiting for dad to finish cooking. She says: "아빠, 배고파요! ë°¥ ë‹€ 됐얎요?" Her appearance references @Image 1. Then the camera pans right to @Image 4's composition. Dad's appearance references @Image 2. Dad replies: "거의 ë‹€ 됐얎, 조ꞈ만 Ʞ닀렀!" Then cut back to a close-up of the daughter's slightly disappointed expression: "아직 멀었얎요? 맛있는 냄새 나는데..." Then cut to dad's close-up: "읎제 진짜 ꞈ방읎알. 빚늬빚늬 하지 말고 손부터 씻고 와!"

04

ビデオ参照

4.1 アクション参照

✹ プロンプトテンプレヌト

Reference "action description" from "Video n", generate "scene description", keep action details consistent.

参考資料

Character image

Character image

Action reference

最終結果

📝 プロンプト

詊しおみる

Reference the character actions and camera language from @Video 1, generate a fight scene with @Image 2 and @Image 1. @Image 2 is the left character, @Image 1 is the right character. With intense background music.

参考資料

Horse running reference

最終結果

📝 プロンプト

詊しおみる

Reference the horse's running form from @Video 1, generate a golden stallion galloping on a grassland, then freeze its magnificent running pose, transforming into a horse-shaped gold pendant.

4.2 カメラ参照

✹ プロンプトテンプレヌト

Reference "camera movement" from "Video n", generate "scene description", keep camera movement consistent.

参考資料

Tech park building

Tech park building

Camera movement reference

最終結果

📝 プロンプト

詊しおみる

Reference the camera movement from @Video 1, create a concept video of a tech park with the high-rise from @Image 1 as the visual center, also using a first-person diving perspective, highlighting the tech feel of the park in @Image 1.

4.3 VFX 参照

✹ プロンプトテンプレヌト

Reference "VFX description" from "Video n", generate "scene description", keep VFX consistent.

参考資料

Flute girl

Flute girl

Golden particle effects

最終結果

📝 プロンプト

詊しおみる

Reference the golden particle effects from @Video 1, have the character in @Image 2 playing the flute while surrounded by the same particle effects.

参考資料

Wings girl

Wings girl

Wing effects reference

最終結果

📝 プロンプト

詊しおみる

Reference the effects from @Video 1, have the girl in @Image 1 grow the same wings, with the wing generation trajectory matching.

05

ビデオ線集

5.1 ゚レメントの修正

✹ プロンプトテンプレヌト

Add elements: Add "ideal element description" at "time position" + "spatial position" in "Video n". Remove elements: Remove "element to delete" from "Video n", keep other content unchanged. Modify elements: Replace "element to replace" in "Video n" with "ideal element description".

参考資料

Original video

最終結果

📝 プロンプト

詊しおみる

Add fried chicken, pizza, and other snacks to the counter in @Video 1.

参考資料

Original video

最終結果

📝 プロンプト

詊しおみる

Clear other parts and tools from the desktop in @Video 1, keep the desktop clean and tidy, leaving only what they're holding.

参考資料

Replacement face cream

Replacement face cream

Original video

最終結果

📝 プロンプト

詊しおみる

Replace the perfume in @Video 1 with the face cream from @Image 1, keeping the motion and camera movement unchanged.

5.2 ビデオを拡匵

✹ プロンプトテンプレヌト

Extend "Video n" forward/backward + "description of extended video" Generate content before/after "Video n" + "description of extended video"

💡 Automatically clips adjacent parts for blending. The original video clip is not repeated.

参考資料

Original video

最終結果

📝 プロンプト

詊しおみる

Generate content after @Video 1: two late-arriving men run towards them, the five finally meet and chat happily.

参考資料

Original video

最終結果

📝 プロンプト

詊しおみる

Extend @Video 1 forward, give the man in white an over-the-shoulder shot. He says: "It's not that bad. You're just stressed. Everyone goes through this, you just need to keep going."

5.3 トラックをスプラむス

✹ プロンプトテンプレヌト

"Video 1" + "transition description" + cut to "Video 2" + "transition description" + cut to "Video 3"

💡 Supports up to 3 video inputs, max 15 seconds. Will automatically clip and blend overlapping sections.

参考資料

Video 1

Video 2

最終結果

📝 プロンプト

詊しおみる

@Video 1, the moment the leaf touches the ground, golden particle effects burst out, a breeze blows through, cut to @Video 2.

06

ベストプラクティス

コア原則

1

具䜓的に

プロンプトが明確で正確であればあるほど、予枬䞍可胜たたは奇劙な結果が埗られる可胜性は䜎くなりたす。

2

少ないほどよい

䞻䜓たたはアクションを修食語で過負荷にしないでください — フォヌカスが曖昧になりたす。

3

参考資料を掻甚

テキストだけでは耇雑な構成、カメラ移動、効果を説明できない堎合は、適切な画像たたはビデオアセットを芋぀けお支揎しおください。

4

物理的劥圓性

物理的に䞍可胜なシナリオの説明は避けおください — モデルはある皋床たで珟実の物理孊に䟝存しおいたす。

💡 普遍的なスタヌタヌフォヌミュラ

[Scene type] + [Subject 1][State 1] + [Subject 2][State 2] + [Environment] + [Lighting & Mood] + [Camera Movement]