Arkscale

A lightweight web application that upscales low-resolution images using AI-based super-resolution models in Python.

This project was built to solve a common problem I encountered while working on side projects: very small images (e.g., 20x20 px icons) becoming unusable when scaled manually. Most free online tools were limited, unreliable, or restricted, so I built a simple, reproducible solution that performs high-quality image upscaling locally.

Tech Stack

Python Docker

Python Libraries

Streamlit (UI layer) Super-Image (super-resolution models)

How It Works

1. User uploads a low-resolution image.
2. The image is processed using a super-resolution model from the super-image package.
3. The AI model reconstructs a higher-resolution version of the image.
4. The upscaled image is displayed and available for download.

The project focuses on making AI upscaling accessible through a simple and clean interface while keeping deployment straightforward.

Local Deployment

1. Clone Repository

• git@github.com:ArkkanElkhatib/Arkscale.git && cd Arkscale

2. Build Docker Container

• docker build -t arkscale .

3. Deploy Container

• docker run -it -p 80:80 arkscale

4. Visit application at http://localhost:80/