Aurasr V2

🚀 AuraSR-v2

A GAN-based super-resolution model for upscaling generated images. It's a variation of the GigaGAN paper for image-conditioned upscaling, with its Torch implementation based on the unofficial lucidrains/gigagan-pytorch repository.

🚀 Quick Start

This model is a GAN-based super-resolution model designed for upscaling generated images. It is a variation of the GigaGAN paper, specifically for image-conditioned upscaling. The Torch implementation is based on the unofficial lucidrains/gigagan-pytorch repository.

📦 Installation

You can install the aura-sr package using the following command:

$ pip install aura-sr

💻 Usage Examples

Basic Usage

from aura_sr import AuraSR

aura_sr = AuraSR.from_pretrained("fal/AuraSR-v2")

Advanced Usage

import requests
from io import BytesIO
from PIL import Image

def load_image_from_url(url):
    response = requests.get(url)
    image_data = BytesIO(response.content)
    return Image.open(image_data)

image = load_image_from_url("https://mingukkang.github.io/GigaGAN/static/images/iguana_output.jpg").resize((256, 256))
upscaled_image = aura_sr.upscale_4x_overlapped(image)

📄 License

This project is licensed under the Apache-2.0 license.