Skip to content

Latest commit

 

History

History
60 lines (46 loc) · 1.94 KB

README.md

File metadata and controls

60 lines (46 loc) · 1.94 KB

Neural Style Transfer Application 🚀

This is a PyQt-based application for performing Neural Style Transfer on images, videos, and live camera feeds. It leverages TensorFlow Hub for pre-trained models and CUDA for accelerated processing on compatible GPUs.

Demo

Demo GIF

Features

  • Image Styling: Apply artistic styles to static images.
  • Video Styling: Process and stylize videos frame by frame efficiently.
  • Live Camera Feed: Real-time style transfer for live camera input.
  • Model Flexibility: Select models from TensorFlow Hub or your system.
  • Interactive UI:
    • Select input images and videos via file dialogs.
    • Preview and process results directly in the application.
    • Output display with zoom for images and playback for videos.
  • Processing Status: Animated popups indicating processing progress.

Installation

Prerequisites

  • Python 3.8+
  • CUDA Toolkit (if using GPU acceleration)
  • Required Python libraries (install via requirements.txt)

Setup

  1. Clone the repository: 
    git clone https://github.com/jayeshrdeotalu/Nest-Neural-Style-Transfer.git   
cd Nest-Neural-Style-Transfer
  2. Install the dependencies: 
      pip install -r requirements.txt
  3. Ensure CUDA is installed and properly configured for GPU acceleration (optional).

Usage

  1. Run the application: 
    python app.py
  2. Select the operation:
  • Style an Image: Choose an input image and a style model.
  • Style a Video: Select a video and apply the chosen style.
  1. Process the selected input:
  • Click on "Process the Styling" to apply the effect.
  • View the results directly in the application.
  1. Save outputs:
  • Output is saved in the same input folder.

License

This project is licensed under the MIT License.

Acknowledgments

  • TensorFlow Hub: For providing pre-trained models.
  • PyQt: For the UI framework.
  • CUDA: For GPU acceleration.