deeptracer 0.1.3

A powerful tool for detecting and analyzing deepfake images and videos.

DeepTracer

DeepTracer is a powerful Python tool designed to detect and analyze deepfake images and videos. Utilizing state-of-the-art deep learning techniques, it provides reliable predictions to identify manipulated media.

Installation

You can install DeepTracer using pip:

pip install deeptracer

Usage

here's how you can use the deeptracer package to detect deepfakes in images and videos.

Import the DeepFakeDetector

To begin using DeepTracer, simply import the DeepFakeDetector class:

from deeptracer import DeepFakeDetector

Image Prediction

You can detect whether an image is a deepfake using the predict_image method. Here's an example:

detector = DeepFakeDetector()
image_result = detector.predict_image('path\image.jpg')
print(image_result)

Video Prediction

Similarly, you can analyze videos for deepfake detection using the predict_video method:

detector = DeepFakeDetector()
video_result = detector.predict_video('path\video.mp4')
print(video_result)

Expected Output

The output of both predict_image and predict_video methods will be in a format that indicates whether the media is likely to be real or fake, along with a confidence score. For example:

{
  "label": "fake",  # or "real"
  "confidence": 0.95  # Confidence score in percentage
}

Requirements

The following dependencies are required and automatically installed when using the package:

• torch
• facenet-pytorch
• Pillow
• opencv-python
• numpy
• gdown

These libraries are critical for the deep learning models and image/video processing.

License

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

Author

Developed by Vishwa. For any inquiries or suggestions, feel free to reach out at [e-mail].