@CycleDecoded: Unbelievable, folks — for those doing film commentary and short video analysis, you can truly kick back now. This divine project I just dug up on GitHub completely automates the grueling work of shot-by-shot analysis. Called Lapian-Notes, it's an AI-powered tool that turns movies into…
Summary
Lapian-Notes is an open-source AI-assisted film analysis tool that automatically extracts frames, generates scene timelines and emotion curves, runs entirely locally, and requires no data uploads.
View Cached Full Text
Cached at: 07/10/26, 06:15 PM
Unbelievable, folks. If you do film commentary or short video breakdowns, this is the one that lets you truly kick back and relax.
Just unearthed this god-tier project on GitHub that completely automates the grueling, all-night slog of film analysis.
The project is called Lapian-Notes (拉片笔记) — an open-source tool that turns any movie into an AI-assisted analysis notebook. You feed it a video, and it breaks the film down into a level of granularity that would even spook the director. Everything runs locally, so no worries about copyright or privacy. It’s hardcore open-source.
Someone actually made this thing. How brutal is it? Check out the goods:
- Local full-auto frame extraction: Say goodbye to slamming the space bar. AI automatically cuts by shot.
- Storyline swimlane timeline: Exactly what happens at what minute/second? It generates a visual chart on the spot.
- Emotion curve fully mapped: Where’s the climax? The low point? Instantly reveal the rhythm secrets of a viral video.
- God’s-eye structure tree: Three-act structure or multi-line narrative? It dissects the skeleton down to your underwear.
- Pure local free running: Freeloaders rejoice — no need to burn money on APIs. Just run it on your own machine.
For anyone in film deconstruction, searching for benchmark viral content, or working on content structuring, this is basically a literal cheat code. If you know, you know.
GitHub link: https://github.com/bkingfilm/lapian-notes
bkingfilm/lapian-notes
Source: https://github.com/bkingfilm/lapian-notes
拉片笔记 (Lapian Notes)
Discord (https://discord.gg/uT6xryBX9w) X (https://x.com/bkingfilm) Download (https://github.com/bkingfilm/lapian-notes/releases/latest) License
Turn a movie into an editable shot-by-shot study notebook — local frame extraction, automatic subtitle generation, one-click AI analysis package creation, import AI results to get storyline swimlane timeline, structure tree, and audience emotion curve.
Designed for creators who want to systematically learn film storytelling. Everything runs locally; film and note data never leaves your machine.
English: Lapian Notes turns a film into an editable shot-by-shot study notebook — local frame extraction, AI-assisted structure analysis (bring your own AI, no API key), story-line swimlanes, structure tree and audience-emotion curve. UI is currently Chinese-only; everything runs locally.
After downloading and extracting: Windows double-click
run.bat, Mac double-clickrun.command. Wait for it to auto-install, and your browser will open the tool automatically.
Screenshot: storyline swimlane timeline, audience emotion curve, structure tree with frame screenshots
What it can do
- Movie timeline: Import a video, extract frames at 1-second intervals, generate a full-film timeline with screenshots.
- AI analysis package: Package screenshots, subtitles, and plot info into a ZIP, send to any AI like ChatGPT for analysis (no AI service binding, no API key required).
- Storyline swimlane chart: After AI returns results, spread out multi-swimlane timelines based on the film’s own storylines (AI-customized names), auto-generate reference cards for overlapping segments.
- Structure tree & emotion curve: Group segments by main line, sub-line, emotional line, information line; audience engagement intensity curve marks the film’s rhythm fluctuations.
- Deep segment breakdown: Any segment can be individually packaged and sent to AI, broken down to scene and shot level (script sections, audiovisual techniques).
- Manual refinement: Each segment has a full editing panel (segment function, key beats, script restoration, creative intent, etc.).
- Export: Export entire notes as Markdown, export script text separately.
Workflow
A persistent four-step guide sits at the top of the interface — just follow along:
- Import movie: Select the video file. Then fully automatic: incompatible formats are auto-transcoded locally, frames extracted, embedded subtitles read (or auto-search online subtitles if missing), and finally auto-generate the AI analysis package.
- Send analysis package to AI: Pass the ZIP to any AI like ChatGPT. First paste the instruction (already copied to your clipboard), then upload the ZIP.
- Import AI results: After AI returns a JSON file, click “Import AI Results” — timeline, structure tree, and emotion curve are auto-generated.
- Fine-tune analysis: Click on a segment to write notes; for key segments, use “Export this segment only to AI” for deep breakdown. When done, export as Markdown.
Quick Start
Don’t know programming? Three steps to use
- Download: Click the green “Code” button on this page, select “Download ZIP”.
- Extract: Double-click or right-click to extract the downloaded archive.
- Launch:
- Windows: Double-click
run.batin the folder. - Mac: Double-click
run.commandin the folder. The first time, the system may say “cannot be opened”. Control-click on the file → Open → Open. This is only needed once; after that, just double-click.
- Windows: Double-click
The first launch automatically prepares the runtime environment and program components (if you don’t have Node.js installed, it will automatically download a portable version — takes a few minutes; subsequent launches are instant). Once done, the tool page opens automatically in your browser. Keep the black service window open while using; close it to exit.
For developers
Requirements: Node.js 18+, Chrome-based browser (Chrome / Edge). ffmpeg (https://ffmpeg.org/) optional — only needed for automatic transcoding of formats like RMVB/AVI/HEVC; H.264 MP4 works without it.
npm install
npm run dev
Open the address shown in the terminal (default http://localhost:5173) in your browser.
Note: Automatic transcoding and automatic subtitle search are provided by the dev server’s local backend. You must run
npm run dev(or the startup script) to have full functionality. When statically deployed (npm run build), these two features fall back to manual operation; all other features remain unaffected.
Data Storage
- Notes text is auto-saved in the browser’s localStorage; frame screenshots are cached in IndexedDB — all on your local machine.
- “Save Project” exports a self-contained ZIP (notes + screenshots + Markdown). Use it to backup or transfer to another browser.
- Due to browser security limitations, after refreshing, you need to re-select the video file to re-extract frames (existing notes and screenshots are unaffected).
Disclaimer
- This tool is intended solely for personal study and research of film storytelling. Use only videos for which you have legal access.
- The online subtitle search feature retrieves subtitles from public subtitle databases. Subtitle copyright belongs to the original creators. Do not use them for any commercial purpose.
Communication & Feedback
- To discuss film analysis, movies, and game visuals: Join the BKinGfilm Discord community (https://discord.gg/uT6xryBX9w) (2500+ members).
- Found a bug or have a feature request: Open an Issue (https://github.com/bkingfilm/lapian-notes/issues).
- Follow tool updates and author’s news: X @bkingfilm (https://x.com/bkingfilm).
License
Similar Articles
@XAMTO_AI: There is an open-source project called narrator-ai-cli-skill. Plug it into an agent like Claude Code, and just say 'Help me make a commentary video for The Shawshank Redemption', and the AI handles everything: automatically generates commentary script without you lifting a finger, precisely matches corresponding movie clips, …
Introducing an open-source project called narrator-ai-cli-skill, which can be integrated into AI agents like Claude Code. With just one sentence, users can automatically complete the entire process of creating a movie commentary video — script, voiceover, editing, background music, etc. — greatly reducing the production cost of film commentary channels.
@iluciddreaming: AI short drama production, the workflow is already running. Toonflow, with 11.6k stars on GitHub: Throw in a script, AI writer breaks it down, intelligent storyboarding generates shot scripts, workflow runs automatically. Free and open source. The most noteworthy point: it actually works, not a demo.
Toonflow is a free and open-source AI tool that automatically breaks down scripts into shot scripts and generates storyboards, with an automated workflow that is truly usable.
@berryxia: Wow! This project is amazing! Another complete AI video creation platform has been open-sourced, with both frontend and backend under the MIT license. Magiviz has turned 'write script → define characters → draw storyboards → generate video → final cut' into a full-process AI workflow, supporting up to 60-second videos with 12 built-in mainstream video models. The author...
Magiviz is an open-source AI video creation platform that provides a full-process AI workflow from writing scripts to generating videos, supporting up to 60-second videos with 12 built-in mainstream models, and the code is open-sourced under the MIT license.
@CycleDecoded: Unbelievable, guys, someone actually made this—it's directly taking away the livelihood of video re-uploaders. Just found an open-source magical tool on GitHub that quietly makes money: Y2A-Auto. This thing is an emotionless automated pipeline that can fully automatically download YouTube videos, process them in one go, and post to …
Y2A-Auto is an open-source, fully automatic video repurposing tool that downloads videos from YouTube, uses AI for translation, subtitle generation, and content moderation, and automatically uploads to Bilibili and AcFun for unattended operation.
@CycleDecoded: Guys, stop hoarding junk in your file assistant and favorites. This insane open-source project just dropped on GitHub turns 'read later' into a fully automatic AI assistant. Freeloaders and info hoarders, rejoice! It's called SuperBrain, built by developer sidinsearch…
SuperBrain, a newly released open-source project on GitHub, is a fully self-hosted AI 'second brain' that supports one-tap saving on Android and AI automatic analysis of videos, posts, etc., using a local SQLite database with zero subscription fees.