分格與分鏡腳本

找找看人群插畫

把任何地點和角色變成尋寶式的密集人群插畫:數百個忙碌小人,你指定的角色就藏在人群裡某處

更新日期:

完整提示詞

英文提示詞包含建議的範例需求。生成前請調整主體和細節。原始範例輸入及生成參數尚未完全核實。

推薦模型: GPT Image 2

Illustrate a giant seek-and-find crowd scene in the classic wimmelbild style: a slightly elevated, gently tilted overhead view of a sprawling location packed edge to edge with hundreds of tiny, fully drawn people, each busy with their own funny little activity — chatting, chasing pets, dropping ice cream, taking photos, hauling packages. Use the place, theme, and target character described in the user request; hide that one target somewhere non-obvious in the crowd, drawn with a slightly distinctive outfit or pose so a patient viewer can eventually spot them, but never centered or spotlighted. Render in a clean, flat, colorful storybook illustration style with crisp linework, simple cel shading, a bright cheerful palette, and consistent miniature proportions so no figure dominates. Layer buildings, vehicles, stalls, and street furniture to create pockets, corridors, and small visual jokes that reward slow scanning. Keep detail sharp everywhere with no depth-of-field blur, and add no text, labels, arrows, watermarks, or logos. User request: a girl in a red striped scarf hidden in a bustling winter market

官方範例

找找看人群插畫
範例模型: Seedream 5.0 Lite
找找看人群插畫
範例模型: Seedream 5.0 Lite

來源與署名

由 OmniArt 改編及重寫,以下列出來源署名、修改說明及使用條件

  • Where Is [Your Name] Crowd Search Poster

    ZeroLu/awesome-gpt-image

    README states CC BY 4.0 while the current LICENSE is MIT; attribution and the umbrella MIT notice are retained in THIRD_PARTY_NOTICES.md.

    Rewritten and consolidated into an OmniArt recipe archetype; template tokens removed; named people, brands, platforms, games, and film IP generalized; no raw prompt text or source image copied.

來源與署名

相關的提示詞配方