RECUT
RECUT is an AI-powered video genre transformer that takes any ordinary video and re-imagines it as a cinematic production in a genre of your choice — Horror, Action, Romance, Comedy, Thriller, or Epic Documentary.
Core trick
CLIP scores sampled frames against genre prompts, then Gemini writes a timeline with voice-over that MoviePy renders with genre filters.
Description
RECUT is an AI-powered video genre transformer that takes any ordinary video and re-imagines it as a cinematic
production in a genre of your choice — Horror, Action, Romance, Comedy, Thriller, or Epic Documentary. The pipeline
uses CLIP (ViT-B-32) for local semantic frame analysis and scoring, Gemini 2.5 Flash as an AI director that crafts a
narrative timeline with voice-over scripts, Edge-TTS for cinematic narration, and MoviePy for rendering with
genre-specific visual effects (color grading, vignettes, film grain, Ken Burns zoom). It also features a "Moment
Finder" — a semantic video search engine where you can type natural language queries like "red car passing" or
"someone smiling" and instantly find the exact moment in any video using CLIP embeddings. The app solves the problem
of making cinematic content creation accessible to anyone — no editing skills needed, just upload a video, pick a
genre, and get a fully narrated short film in under 90 seconds.
README excerpt
# RECUT - Change the Genre of Reality RECUT is an AI-powered video genre transformer that takes any ordinary video and re-imagines it as a cinematic production in the genre of your choice. Upload a video, pick a genre, and get a fully narrated short film in under 90 seconds. It also features **Moment Finder** — a semantic video search engine where you type natural language queries like *"red car passing"* or *"someone smiling"* and instantly find the exact moment in any video. ## How It Works ``` Video → CLIP (frame analysis) → Gemini (AI director) → Edge-TTS (narration) → MoviePy (render) → Cinema ``` 1. **Frame Extraction** — OpenCV extracts keyframes, histogram-based deduplication removes redundancy 2. **CLIP Analysis** — ViT-B-32 scores frames against genre-specific prompts and generates semantic descriptions 3. **Gemini Director** — Gemini 2.5 Flash receives a mosaic of top frames + descriptions and writes a full narrative timeline with voice-over scripts, shot durations, and visual effects 4. **Voice-Over** — Edge-TTS generates cinematic narration with genre-appropriate voices and pacing 5. **Rendering** — MoviePy composites everything with genre-specific color grading, vignettes, film grain, and Ken Burns zoom effects ### Moment Finder - Encodes every frame of a video with CLIP embeddings (3 fps) - Accepts natural language search queries - Returns the top matching moments with timestamps, thumbnails, and confidence scores - Click any result to jump directly to t