Face Detector

Face Detector is a Python-based tool that performs face detection in videos and identifies specific scenes where a given face appears.

Features

• Detect faces in video frames using a reference image.
• Extract specific scenes from a video where the given face is detected.
• Automatically generate new video files with detected scenes.
• Designed for future extension with a UI.

Installation

Prerequisites

• Python 3.8+
• ffmpeg installed on your system:
  • For Ubuntu/Debian: sudo apt install ffmpeg
  • For macOS: brew install ffmpeg
  • For Windows: Download and install from ffmpeg.org.

Clone the Repository

git clone https://github.com/katanabana/face-detector-project.git
cd face-detector-project

Install Dependencies

pip install -r requirements.txt

Usage

Here's an example of how to use the FaceDetector class in your own script:

from face_detector.face_detector import FaceDetector

video_path = 'video.mp4'
verification_image = 'face.png'
output_directory = 'result'

face_detector = FaceDetector(video_path, verification_image, callback=print)

# Adjust different parameters of processing
scenes = face_detector.relevant_scenes(tolerance=0.7, frequency=10, quality=0.5)

face_detector.write(scenes, output_directory)

To test the module with sample data:

1. open terminal and navigate to the root directory of the cloned repository of this project
2. run in terminal: python -m example.example
3. check if in example folder was created a directory result containing output files of scenes from example/video.mp4 where the face on example/face.png appears.

Future Plans

• UI Development: We plan to add a graphical user interface to make this tool more accessible to non-developers.
• Additional Features: Advanced face matching, add option to detect specific moments instead of scenes, multiple faces detection, etc.