https://github.com/albertomarquillas/face-anonymizer
Privacy-focused face anonymization tool using OpenCV and MediaPipe with blur, pixelate, and emoji modes. Portfolio-ready with CLI, docs, and tests.
https://github.com/albertomarquillas/face-anonymizer
cli-tool computer-vision face-detection image-processing mediapipe opencv privacy python
Last synced: about 2 months ago
JSON representation
Privacy-focused face anonymization tool using OpenCV and MediaPipe with blur, pixelate, and emoji modes. Portfolio-ready with CLI, docs, and tests.
- Host: GitHub
- URL: https://github.com/albertomarquillas/face-anonymizer
- Owner: AlbertoMarquillas
- License: mit
- Created: 2025-09-07T02:29:50.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-09-07T22:06:49.000Z (10 months ago)
- Last Synced: 2025-09-07T23:32:37.102Z (10 months ago)
- Topics: cli-tool, computer-vision, face-detection, image-processing, mediapipe, opencv, privacy, python
- Language: Python
- Homepage:
- Size: 8.79 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Face Anonymizer
## Overview
Face Anonymizer is a privacy‑focused tool that automatically detects and anonymizes faces in images, videos, or webcam streams. It supports multiple obfuscation modes such as **blur**, **pixelate**, and **emoji overlay**. The project is designed as a portfolio‑ready demo with a clean CLI, documentation, and tests.
## Repository Structure
```
face-anonymizer/
├─ src/ # Source code (main CLI)
├─ test/ # Tests and examples
├─ docs/ # Documentation
│ └─ assets/ # Images, diagrams
├─ models/ # Model files (not tracked)
├─ data/ # Local datasets (not tracked)
├─ notebooks/ # Jupyter notebooks (optional)
├─ build/ # Output files (not tracked)
├─ configs/ # Config files (YAML)
├─ archive/ # Legacy code
├─ requirements.txt # Minimal dependencies
├─ LICENSE # MIT License
└─ README.md # Project description
```
## Getting Started
### Prerequisites
* Python 3.10+
* Install dependencies:
```powershell
pip install -r requirements.txt
```
### Usage
#### Anonymize an Image
```powershell
python .\src\main.py --input .\data\your_image.jpg --method blur --output .\build\out.jpg
```
#### Anonymize a Video
```powershell
python .\src\main.py --input .\data\your_video.mp4 --method pixelate --output .\build\out_video.mp4
```
#### Anonymize from Webcam
```powershell
python .\src\main.py --webcam 0 --method blur --draw --output .\build\out_cam.mp4
```
#### Use Emoji Overlay
```powershell
python .\src\main.py --input .\data\your_image.jpg --method emoji --emoji .\data\emoji.png --output .\build\emoji_image.png
```
### CLI Options
* `--input / -i`: Path to image or video.
* `--webcam / -w`: Webcam index (default 0).
* `--method / -m`: Anonymization method (`blur`, `pixelate`, `emoji`).
* `--emoji`: Path to PNG emoji (required for `--method emoji`).
* `--draw`: Draw bounding boxes around detected faces.
* `--output / -o`: Output path for anonymized media.
* `--config / -c`: YAML config file for detector settings.
## Dataset & Models
* Place your own media in `data/`. See [data/README.md](data/README.md).
* Place pretrained models or cascade files in `models/` (not tracked). Add instructions in `models/README.md`.
## Features
* Real‑time face detection using MediaPipe.
* Multiple anonymization modes (blur, pixelate, emoji overlay).
* Simple CLI with PowerShell examples.
* Modular codebase with configs and archive for legacy code.
## What I Learned
* Practical integration of MediaPipe FaceDetection with OpenCV.
* Building a CLI‑based tool with relative paths and configs.
* Managing datasets/models in a portfolio‑ready structure.
* Using conventional commits and clear repository scaffolding.
## Roadmap
* [ ] Add test suite in `test/`.
* [ ] Expand anonymization methods (mosaic, cartoon).
* [ ] Dockerfile for containerized usage.
* [ ] Optional GPU acceleration.
## License
This project is licensed under the MIT License – see the [LICENSE](LICENSE) file for details.