Petrify
ActiveFreeze the world. Keep the clock ticking.
An open source computer vision pipeline that removes people from security footage while preserving the timestamp overlay. When a person enters the scene, the background freezes into a clean still while only the timestamp keeps updating in real time. Frame-perfect, zero-config, fully local.
What it does
Petrify watches the frame for people and, the moment one appears, swaps the live background for a clean pre-selected still. The timestamp region is detected automatically and continues ticking in real time, so the footage stays verifiably current without ever showing a person. Everything runs on your machine.
Key features
- Frame-perfect person detection and removal.
- Automatic timestamp region detection, no configuration needed.
- Fully local processing with no cloud dependencies.
- Three-detector pipeline: MOG2 motion, HOG person detection, static background differ.
- Hardware acceleration on Apple Silicon and NVIDIA GPUs.
- Automatic background-frame selection.
Stack & status
Python 3.10+ with OpenCV, FFmpeg, NumPy and tqdm. Processes 1080p at roughly real time on Apple M-series chips, faster at lower resolutions. Apache 2.0 licensed, in active development.
Use responsibly
Always ensure you have the legal right to process any footage you use. Petrify is built to balance security with privacy, not to circumvent either.