Web Interface for VisionConverter

This project is a web interface for the library VisionConverter.

Overview

This web interface provides an easy-to-use platform for converting and processing datasets using the VisionConverter Python library. It is designed to simplify data handling tasks for users working with various dataset formats in computer vision projects.

Features

• User-friendly web interface for uploading and converting datasets.
• Built with FastAPI, ensuring fast and reliable backend performance for file processing and conversion tasks.
• Supports multiple data formats: YOLO, COCO, Pascal VOC, LabelMe JSON, TensorFlow CSV, VGG Image Annotator JSON and CreateML.
• Example datasets for each format.

Installation

Requirements

• Python 3.12+
• The VisionConverter library (installed automatically via requirements)

1. Clone the repository

    git clone https://github.com/GCousido/TFG-WebInterface.git
    cd TFG-WebInterface

2. Install dependencies

    pip install -r requirements.txt

How to Use

Start the API

    cd API
    python api.py

The web interface will be available at http://localhost:8080.

Interface Usage

1. Upload your dataset files directly through the web interface.
2. Select the desired input and output formats.
3. Start the conversion process.
4. Download the converted files once completed.

License

This project is licensed under the MIT License.

---

For more details on the underlying conversion capabilities, refer to the VisionConverter library documentation.