Skip to content

Latest commit

 

History

History
52 lines (34 loc) · 1.57 KB

README.md

File metadata and controls

52 lines (34 loc) · 1.57 KB
Raw

Image Segmentation and Tagging

SegTag (segtag.py) is a script designed to segregate and classify images based on various criteria such as facial features, body posture, and more. It leverages computer vision techniques and the Gradio interface to provide an interactive GUI for users.

Features

  • Angle Segregator: Classifies images based on the angle of the shot (e.g., close-up, portrait, full body).
  • Hairstyle Segregator: Uses the Roboflow API to classify hairstyles in images (integration in progress).
  • Clothing Segregator: Placeholder function for future development.
  • WD14 Tagging: Placeholder function for future development.

Prerequisites

  • Python 3.x
  • Gradio v3.44.3
  • OpenCV
  • Mediapipe
  • Roboflow (for Hairstyle Segregator)

Installation

  1. Clone the repository:

    git clone <repository-url>

  2. Navigate to the directory:

    cd path-to-directory

  3. Install the required packages:

    pip install -r requirements.txt

  4. Run the script:

    python segtag.py

  5. Open the Gradio interface in your browser using the provided link.

  6. Use the GUI to select the desired functionality and provide the necessary inputs.

Easy Install

Just run the setup (setup.bat) file in Windows and let that take care of building your venv and downloading the dependencies for you. To initialise the UI you can then use the run-segtag (run-segtag.bat) file.

Contributing

Contributions are welcome! Please create a pull request with your changes.

License

This project is licensed under the MIT License.