Skip to content

README.md

Latest commit

 

History

History
238 lines (167 loc) · 3.36 KB

README.md

File metadata and controls

238 lines (167 loc) · 3.36 KB
Raw

🧠 NeuralCanvas

Transform Reality Into Art with AI

NeuralCanvas is a Dockerized AI-powered artistic style transfer platform built using Adaptive Instance Normalization (AdaIN) and PyTorch. The application allows users to blend the content of one image with the artistic style of another, generating unique AI artwork in seconds through an intuitive web interface.

🚀 Live Demo

Web Application: https://neuralcanvas-cpk1.onrender.com

GitHub Repository: https://github.com/itsakki10/NeuralCanvas

✨ Key Features

  • 🎨 AI-powered Neural Style Transfer
  • 🧠 Adaptive Instance Normalization (AdaIN)
  • 📤 Upload custom content and style images
  • 🎚 Adjustable style intensity control
  • ⚡ Real-time image generation
  • 👀 Live image preview
  • 📥 Download generated artwork
  • 🖼 Interactive example gallery
  • 📱 Fully responsive user interface
  • 🐳 Docker container support
  • 🌐 Production-ready deployment with Gunicorn

🖼 How It Works

Content Image
      ↓
Style Image
      ↓
AdaIN Encoder
      ↓
Feature Alignment
      ↓
Decoder Network
      ↓
Stylized Artwork

📸 Screenshots

Homepage

Homepage

Style Transfer Studio

Upload Studio

Generated Artwork

Generated Result

🏗 System Architecture

User Uploads Images
          ↓
      Flask App
          ↓
     Image Preprocessing
          ↓
      AdaIN Network
     (PyTorch Model)
          ↓
    Style Transfer Engine
          ↓
   Generated Artwork
          ↓
      Browser Output

🛠 Tech Stack

Frontend

  • HTML5
  • CSS3
  • Bootstrap
  • JavaScript

Backend

  • Flask
  • Flask-WTF
  • Gunicorn

Artificial Intelligence

  • PyTorch
  • Adaptive Instance Normalization (AdaIN)
  • VGG Encoder
  • Decoder Network

DevOps & Deployment

  • Docker
  • Render

📂 Project Structure

NeuralCanvas/
│
├── AdaIN/
├── assets/
├── content_data/
├── experiment/
│   └── big_dataset/
├── static/
├── style_data/
├── templates/
├── utils/
│
├── app.py
├── train.py
├── requirements.txt
├── Dockerfile
├── Procfile
├── runtime.txt
├── README.md
└── vgg_normalised.pth

⚙ Local Installation

Clone Repository

git clone https://github.com/itsakki10/NeuralCanvas.git
cd NeuralCanvas

Create Virtual Environment

python -m venv env

Activate Environment

Windows:

env\Scripts\activate

Linux / macOS:

source env/bin/activate

Install Dependencies

pip install -r requirements.txt

Run Application

python app.py

Open:

http://localhost:5000

🐳 Docker Deployment

Build Docker Image

docker build -t neuralcanvas .

Run Docker Container

docker run -p 5000:5000 neuralcanvas

Open:

http://localhost:5000

📈 Future Enhancements

  • Multiple artistic style blending
  • Batch image processing
  • User authentication
  • Artwork history management
  • Cloud storage integration
  • GPU acceleration support

👨‍💻 Author

Akash Mehra

B.Tech Artificial Intelligence & Machine Learning

Connect With Me

GitHub: https://github.com/itsakki10