파란 붓터치와 회색조로 재연출된 진행자

결과

원작 게시물에서 영상 보기 ↗

제작 · ozansihay · 영상과 장면의 저작권은 제작자에게 있으며, 원작 주소에서 불러옵니다. 원작 보기

Analyze the reference image I uploaded. Write only the final prompt that I can paste into another video editing/generation model.
No video generation, video editing, image generation, or any output generation. No writing descriptions, analysis summaries, titles, bullet lists, or comments. Your response should consist solely of a single ready-to-use prompt text.
The prompt you prepare should be written in a way that can be understood by another model without access to the reference image. Therefore, avoid vague phrases in the prompt such as "like in the reference image," "the style in the uploaded image," or "this design."
Describe the design language of the reference image clearly and in detail within the prompt: color palette, composition, background texture, typography style, graphic elements, collage/doodle/sticker/poster/editorial details, lighting-contrast, grain, halftone, paper texture, sense of motion, and overall atmosphere.
Specify in the prompt: The person's face, identity, voice, speech, lip sync, facial expressions, and hand movements in the video must be preserved. Only the visual design language should change.
Analyze the topic being narrated in the video. Symbols, icons, shapes, stickers, collage pieces, doodle lines, brush effects, or motion graphic details appropriate to the topic can be added.
From the topic narrated in the video, select only a few key words and use them as short text elements suitable for the design. Do not turn full sentences, long explanations, subtitles, or the entire speech into text. The texts should be few in number, for emphasis purposes, and suitable for the design style.
The final prompt should be clear, applicable, and sufficient on its own for editing a vertical video that is suitable for social media, modern, creative, and premium-looking.

모음 출처 · github.com