flux3 video
定價登入/註冊
黑色星期五 · 最高折扣 30%
10天01小時02分鐘36秒
查看優惠 →

FLUX 影片HOT提示展示

AI 生成

AI 影片AI 圖片

創意工具

模型

Veo 人工智慧影片Kling AI 影片Banana AI 圖片

素材庫我的最愛
升級方案
flux3 video
返回博客

South Park 角色:分鏡取景的視覺參考

運用 12 個 South Park 角色插圖參考、鏡頭大小指引,以及在不同畫面間保留角色辨識度的實用方法,規劃更清晰的分鏡畫面。

發布日期 2026年10月10日flux3video.dev

本頁更新動態

  • 2026年10月10日 — 文章首次發布。
South Park 角色:分鏡取景的視覺參考

A storyboard needs to make the subject readable before it can explain an action. South Park characters offer useful examples because their identities are carried by large, simple features: a hood, an earflap hat, a wide coat, or a distinctive hairstyle. The camera can move closer, but each crop should retain the clues that the viewer needs. For a new character for a storyboard, try South Park Character Creator.

The artwork in this guide comes from the South Park Public Library. We use it to discuss full figures, medium framing, and close portraits. These are observations and proposed storyboard choices; the images are not video frames generated for this article, and the suggestions are not presented as tested sequences.

Match the frame to the information

A full figure explains posture and costume. A medium view emphasizes the face, clothing, and any hand-level action. A close portrait may communicate identity but remove the supporting details. Choose the frame that answers the scene's question instead of using the same crop for every character.

Eric Cartman: allow for body width

Eric Cartman wearing a turquoise and yellow cap, a broad red jacket, and yellow mittens

Cartman's red coat makes a broad lower shape beneath the turquoise cap. A full-body frame shows that proportion clearly. In a group storyboard, give him enough horizontal space; narrowing the crop to match another boy's cell can make the portrait's most distinctive proportion harder to recognize.

Stan Marsh: protect the top of the hat

Stan Marsh wearing a blue cap with red trim and pom-pom, a brown jacket, and red gloves

Stan's red pom-pom and hat band sit above a brown jacket. They are useful identity clues in a medium view. A closer frame should leave space above the head, because trimming the pom-pom may remove a more informative detail than the extra coat fabric the crop retains below.

Kyle Broflovski: retain the side outline

Kyle Broflovski wearing a green earflap hat and orange jacket

Kyle's green earflaps extend beside his face, while the orange jacket anchors the lower portion. A narrow close-up can lose those side shapes. For a storyboard identification frame, preserve both flaps when possible and use the coat as a secondary clue rather than relying on the eyes alone.

Kenny McCormick: frame the hood opening

Kenny McCormick wearing an orange hood with a dark face opening and brown gloves

Kenny's eyes sit inside the dark opening of his orange hood. The complete enclosure matters to the portrait. A close frame should show how the hood surrounds the face; enlarging only the visible eyes may produce an abstract detail that no longer explains who the viewer is seeing.

Compare where the outline changes

Wendy Testaburger

Wendy Testaburger wearing a pink beret, long dark hair, and a purple jacket

Wendy's long dark hair runs beside the purple jacket beneath a pink beret. A medium frame can show the relationship among all three. If a storyboard only needs a headshot, label it accordingly; the crop will not provide the same costume information as this complete reference.

Bebe Stevens

Bebe Stevens with full blond hair around her head and shoulders and a red top

Bebe's blond hair spreads around the head and shoulders, creating a wider border than Wendy's. The red top remains a large lower area. Leave room at the sides of the frame when the hair shape is part of the scene's recognition cue, rather than cutting it to fit a fixed template.

Butters Stotch

Butters Stotch with a small blond hair tuft and turquoise shirt

Butters' small blond tuft and turquoise shirt make a relatively uncluttered figure. A medium view can preserve both without much empty space. The image is useful for testing whether a storyboard's labels are doing too much work: his distinctive features should remain visible even when the caption is hidden.

Tweek Tweak

Tweek Tweak with outward-pointing blond hair and an olive shirt with an uneven closure

Tweek's outward-pointing hair reaches beyond the face's round shape. The olive shirt adds an uneven central line. A tight crop should not trim all the hair points, since they explain much of the outline. This is a framing observation, not a claim about movement in a particular shot.

Some subjects need the full figure

Jimmy Valmer

Jimmy Valmer wearing a yellow top with crutches visible on both sides

Jimmy's crutches lie on both sides of his yellow top and extend toward the ground. A full-body view communicates their relation to his posture. If an action involves moving across a frame, the reference should retain that arrangement instead of treating a face crop as sufficient planning information.

Timmy Burch

Timmy Burch with orange hair and a red shirt, seated in his wheelchair

Timmy's seated position and wheelchair define the overall space he occupies. The red shirt is only one part of the image. A wide reference allows the wheels, foot area, and handles to remain visible, giving a storyboard more reliable information than an enlarged head-and-shoulders view.

Ike Broflovski

Ike Broflovski with a dark hair tuft, a broad head, and a green top with a sailboat

Ike's large head and small body create a proportion that differs from the older children. In a group frame, avoid enlarging his body to equal theirs. A reference should preserve the source relationship, with the dark hair tuft and separated mouth line still visible at the chosen scale.

Randy Marsh

Randy Marsh in a farmer outfit with a brimmed hat, plaid shirt, and dark outer layer

Randy appears in a farmer costume with a brimmed hat, plaid shirt, and dark outer layer. The full image identifies the outfit variation; a face crop mainly identifies the man. Decide which information the storyboard needs, and name the costume if the planned scene depends on this reference.

Plan three shots with one purpose each

For an original cartoon scene, a simple three-shot plan can establish a figure, show an action, and reveal a response. Imagine a student in a rust-colored raincoat finding a lost library card. The first shot shows the figure at a doorway. The second brings the hand and card closer. The third returns to a readable face and body as the student notices the owner's name.

This is a scene proposal, not an observed generation result. Its value is the division of information: the first shot says where, the second says what, and the third explains the response. Keep the same coat, hairstyle, and prop between shots so viewers do not have to re-identify the subject every time.

A South Park character creator can be part of planning an original still reference, while a South Park character maker brief can describe the desired flat shapes and proportions. Video work requires additional decisions about motion and framing. Review actual outputs for continuity rather than assuming that repeating a character name will preserve every detail.

You can develop a scene idea from the homepage or read other visual-planning topics in the blog. The gallery supports creative observation; it does not establish which model will produce a particular character or action. Choose a character whose outline will remain clear in your chosen frame when you try South Park Character Creator.

Check the cut between two frames

Place two proposed storyboard frames beside one another and identify the information that connects them. The coat color, hat shape, and location of the main prop can help viewers understand that the second image continues the same moment. If all three change, the cut may suggest a new scene even when that was not your intention.

For the lost-library-card example, the first frame can show the card near the student's shoe, while the second includes the hand lifting it. The closer view should make the object clearer, not introduce a different object without explanation. Keep the direction of attention understandable as well: the student looks down before the camera moves toward the ground. A storyboard does not need elaborate artwork to establish these relationships. It needs enough visible information for someone else to follow the action in the same order you intended.

Distinguish Warcraft imagery from ordinary portraits

Stan, Kyle, Cartman, and Kenny with altered appearances at computers in the computer lab in Make Love, Not Warcraft

Make Love, Not Warcraft is the episode associated with South Park World of Warcraft. A powerful player repeatedly kills the boys' game characters, and Cartman leads a plan to build their strength. The story gives the familiar cast a particular gaming context.

The computer-lab frame shows how an ensemble image can establish that context at once. The room and screens explain the activity, while familiar clothing identifies the changed figures. Kyle wears green headwear, Kenny retains the orange hood, and Stan has the blue-and-red cap. Cartman appears in red without his usual hat.

若要製作《南方四賤客》的分鏡腳本,請將這張寬幅畫面中的資訊與標準的 Cartman 南方四賤客肖像進行比較。肖像只呈現身分;整體畫面則將身分與場景及共同行動連結起來。請選擇能回答鏡頭實際問題的參考資料。若要為分鏡腳本發展新角色,試試南方四賤客角色產生器。

flux3 video

以影片為先的創意工作空間,將提示詞和圖片變成視覺故事。

support@flux3video.dev ↗
InstagramXYouTubeTikTokPinterest

產品展示

  • 創建
  • 定價
  • 部落格
  • 開發者 API
  • MCP 伺服器

AI 圖片

  • AI 圖片
  • 影片概念靜態影像的 AI 換臉生成器
  • 南方四賤客角色建立器:發展卡通故事影格
  • Nano Banana AI 圖像產生器
  • FLUX AI 圖像產生器
  • GPT Image 2.6 AI 影像產生器:提示詞、參考圖與修訂
更多 (5)
  • Grok Imagine 圖像產生器
  • MiniMax H3 Image:室內規劃靜態圖
  • Seedream 5.5 指南:準備相機產品靜態圖
  • 玩具車變真車:打造電影感靜態影像
  • 免費線上拼接圖片製作社群貼文

AI 影片

  • AI 影片
  • AI 萬聖節服裝改造指南
  • Bigfoot 影片生成器操作導覽
  • 貓咪騎上 Ultra Bee
  • 聖誕舞蹈潮流:打造溫馨木屋雙人舞影片
  • Doja Cat Juicy AI 舞蹈:編導一場動作研究
更多 (14)
  • 地球縮放產生器:電影感序列
  • Elf On The Shelf AI Video:簡易玩具照片指南
  • Flamingo Floatie AI:平靜的頂樓泳池場景
  • Genjutsu AI 影片產生器,用於影片概念迭代
  • 告別 2026 歡迎 2027 影片指南:規劃清晰的年份揭曉
  • AI 萬聖節舞蹈影片產生器:實用指南
  • Kling AI 影片生成器
  • Like Jennie AI Dance Trend:肖像轉影片指南
  • Ozil Goal AI 影片產生器
  • Seedance 2.0 AI 影片生成器
  • Toc Toc AI 舞蹈生成器
  • Veo 3.1 AI 影片生成器
  • 用於表演指導的 Video to Video AI
  • Hotel Lobby AI 用於你片段中的表演鏡頭

說明

  • 聯絡

AI 資源

llms.txt完整指南總覽功能定價常見問題
向 AI 詢問 flux3 videoChatGPT ↗Claude ↗Perplexity ↗Gemini ↗Grok ↗
© 2026 flux3 video. 保留所有權利。
Sitemap服務條款隱私權政策退款政策

Cookie 設定

我們使用 Cookie。