JuneYaooo/rea-video

逆向一切视频:实拍口播、桌拍、AI 视频、3D、代码动效、混剪,逐镜拆出画面、花字、剪辑和声音是怎么做的,并给出复刻方案。Reverse-engineer any video, shot by shot.

Open original source ↗

From the publisher

README & documentation

Source preview

Read the project’s overview, installation instructions and usage examples. The original README is the source of truth.

逆向一切视频。 实拍口播、桌拍、AI 视频、3D、代码动效、混剪都能拆, 逐个镜头告诉你画面、花字、剪辑和声音是怎么做的,再告诉你怎么照着做一条。 中文 | English | 日本語 [](https://github.com/JuneYaooo/rea-video/stargazers) 左边是 TikTok 上的手机实拍原片,右边是别人用 AI 换了产品重做的版本。 rea-video 判断:左边全部实拍,一个场景、两个机位就能拍完;右边全部 AI 生成,8 个镜头可以合并成 3 次生成。 除了画面怎么做的,每条视频还会拆: rea-video 是给 AI 助手用的。在 Claude Code、Codex 这类能运行命令的 agent 里说: agent 会自己装好命令行工具、读完拆解流程,然后从头跑完。电脑上需要有 ffmpeg 和 Python 3.10 以上。 跑完你会拿到: 配一个看视频的模型,Gemini 豆包都行,把接口地址和密钥告诉 agent 就行。配上之后实拍、AI、3D 分得更准,每个镜头还会反推出提示词。不配也能用,agent 自己能看图的话,会补看没把握的镜头。 混合片 · 19 秒 · 10 镜 AI 人物对话 + 代码界面 + 8 秒真实录屏。5 个对话镜头被界面隔开,但同一个场景,一次生成就够。 视频 · 报告 3D 科普片 · 67 秒 · 21 镜 全部 3D 渲染,AI 旁白。可以合并成 10 次视频生成,也可以让大模型写 Blender / Three.js 代码。 视频 · 报告 产品宣传片 · 70 秒 · 12 镜 83% 是代码动效,12% 录屏。不需要视频模型,写一个 Remotion 工程就能做。 视频 · 报告 AI 音乐 MV · 165 秒 · 86 镜 判出歌是 AI 生成的(实际是 Suno)。看图模型说是「影视剧实拍」的镜头,agent 从奏折上的伪汉字认出是 AI。 视频 · 报告 实拍开箱带货 · 72 秒 · 29 镜 全部实拍:头顶俯拍支架、一块地毯一个机位,无口播,同期声 + 曲库 BGM。转码后工具把 35% 判成 AI,agent 看图复核后 97% 实拍。 视频 · 报告 AI 与真人混合 · 58 秒 · 21 镜 实拍 75%(口播 + 拍手机屏幕演示)+…

JuneYaooo/rea-video on GitHub A short preview, not the full document.

Read the full README ↗ · Preview checked 2026-11-04T13:15:55.478Z

Inside the original README — Document outline
  1. rea-video
  2. 能拆哪些视频
  3. 快速开始
  4. 案例
  5. 复刻效果
  6. 准不准
  7. 怎么判断的
  8. 常见问题
  9. 文档
  10. 社区
  11. 致谢

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Links from the README

References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.

Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

What this repository does

逆向一切视频:实拍口播、桌拍、AI 视频、3D、代码动效、混剪,逐镜拆出画面、花字、剪辑和声音是怎么做的,并给出复刻方案。Reverse-engineer any video, shot by shot.

Repository facts

Owner
JuneYaooo
Primary language
Python
Stars
333
Forks
61
Open issues + pull requests
0
License
Not reported — inspect the license file
Archived
No
Default branch
main
Created
2026-10-29T16:15:01.000Z
Last push
2026-11-04T12:16:53.000Z

Topics and intended use

Owner-supplied topics: agent-skills, ai-video, aigc-detection, claude-code, codex, douyin, ffmpeg, prompt-engineering, remotion, reverse-engineering, short-video, shot-detection, subtitles, tiktok, video-analysis, video-breakdown, video-editing, video-generation, vlm

Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.

Evaluate before installing

Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.

README and project files

Issues and maintenance discussion

Releases and changelog

Source and freshness

Source: GitHub. Metadata observed 2026-11-04T12:18:51.791Z. Daily imports are snapshots, not real-time monitoring.

Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.

Open original source

Related repositories

public-apis/public-apis

EbookFoundation/free-programming-books

donnemartin/system-design-primer

vinta/awesome-python

practical-tutorials/project-based-learning

NousResearch/hermes-agent