1. Project Overview

FuturoGen - YOLOv8 Web Platform

A Python Flask-based web application for image annotation, dataset management, model training, and model export using the YOLOv8 object detection model, developed by Boyan.

2. Open Source License

Licensed under Apache-2.0

3. Screenshots

Screenshots

4. Feature Overview

FuturoGen - YOLOv8 Web Platform is a YOLOv8-based object detection web platform that provides an all-in-one solution covering data management, image annotation, model training, and model export.

Functional characteristics

1. Project Management
   Create, list, and delete projects
2. Image Management

• Upload single or multiple images
• Batch upload images via ZIP files
• Drag-and-drop image uploads

3. Image annotation

• Web-based bounding box annotation
• Navigate between images (previous/next)
• Manage project labels (categories): add, edit, and delete
• Annotation data is saved to the database in real time, with YOLOv8 .txt annotation files generated automatically

4. Dataset splitting
   Batch assign images to training, validation, test, or unassigned sets
5. Model training

• Actually invokes the ultralytics library to train models
• Training tasks run in background threads, while the frontend displays training logs and progress in real time
• Automatically prepares the required dataset directory structure and data.yaml file for training

6. Model export
   Export trained models into multiple formats, including ONNX and TorchScript

5. Tech Stack

Tech Stack

• Backend
  Python 3.9+, Flask, Flask-SQLAlchemy, threading, OpenCV (cv2), Pillow (PIL.Image), ultralytics, python-dotenv, PyYAML
• Frontend
  HTML5, CSS3, JavaScript, Bootstrap 5, Jinja2, Fabric.js
• Database
  SQLite
• Deployment
  Run directly using python app.py

Project Structure


yolov8_web_platform/
├── app.py                 # Flask application entry point
├── requirements.txt       # Project dependencies
├── .env                   # Environment variable configuration
├── models.py              # Database models
├── routes.py              # Route handlers
├── README.md              # Project documentation
├── static/                # Static files directory
│   ├── uploads/           # Uploaded image storage directory
│   ├── datasets/          # Dataset directory
│   └── models/            # Model storage directory
└── templates/             # HTML templates directory
    ├── base.html          # Base template
    ├── index.html         # Home page
    ├── project_detail.html # Project details page
    ├── images.html        # Image management page
    ├── annotate.html      # Image annotation page
    ├── dataset.html       # Dataset splitting page
    ├── train.html         # Model training page
    └── export.html        # Model export page

Installation and Setup

pip install -r requirements.txt

Visit [http://localhost:5005](http://localhost:5005) in your browser