facetrackr 0.2

No project description provided

facetrackr 🕵️‍♂️🎥

A Face Recognition and Timestamp Extraction Tool for Videos

Overview

facetrackr is a Python package that allows users to detect and recognize faces in videos using AWS Rekognition. It provides a seamless way to upload face images to a collection, process videos, and retrieve timestamps where the face is detected.

Features

✅ Add a face image to an AWS Rekognition collection
✅ Process a video to detect the given face
✅ Retrieve timestamps of face appearances
✅ Easy-to-use command-line interface

---

Installation

Ensure you have Python 3.7+ installed. Then install facetrackr using:

pip install facetrackr

---

Usage

After installing, run the following command in your terminal:

python -m facetrackr

You will be prompted to enter:

1. The path of the face image you want to recognize.
2. The path of the video in which you want to detect the face.

Example

Enter the path of the face image: ./images/person.jpg
Enter the path of the video: ./videos/sample.mp4

If the face is found in the video, the output will be:

Starting face detection in video...

Final Detection Report:
Face detected at 12.5 seconds with ID: person_123
Face detected at 28.7 seconds with ID: person_123

---

Demo Code

You can also integrate facetrackr into your own Python script. Below is an example demonstrating how to use the package:

from facetrackr.face_collection import FaceCollection
from facetrackr.video_processing import VideoProcessor

def main():
    face_image_path = input("Enter the path of the face image: ").strip()
    video_path = input("Enter the path of the video: ").strip()

    # Initialize face collection handler
    face_collection = FaceCollection()
    
    # Add face to AWS Rekognition collection
    face_id, external_image_id = face_collection.add_face(face_image_path)

    if face_id:
        print("\nStarting face detection in video...\n")

        # Initialize video processor
        video_processor = VideoProcessor()
        detection_results = video_processor.process_video(video_path, external_image_id)

        print("\nFinal Detection Report:")
        for detection in detection_results:
            print(f"Face detected at {detection['timestamp']} seconds with ID: {detection['external_image_id']}")
    else:
        print("Face was not added to collection. Exiting.")

if __name__ == "__main__":
    main()

---

Requirements

Before running the package, install dependencies:

pip install -r requirements.txt

---

AWS Configuration

Since this package relies on AWS Rekognition, ensure you have:

1. An AWS account with Rekognition enabled.
2. AWS credentials configured using:

   aws configure
   

3. The necessary permissions in IAM for Rekognition.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---