Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/fal-ai/aura-sr
AuraSR: GAN-based Super-Resolution for real-world
https://github.com/fal-ai/aura-sr
Last synced: about 11 hours ago
JSON representation
AuraSR: GAN-based Super-Resolution for real-world
- Host: GitHub
- URL: https://github.com/fal-ai/aura-sr
- Owner: fal-ai
- License: cc-by-sa-4.0
- Created: 2024-06-25T22:20:37.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2024-11-13T19:40:00.000Z (about 1 month ago)
- Last Synced: 2024-12-15T22:43:39.755Z (8 days ago)
- Language: Python
- Size: 28.3 KB
- Stars: 414
- Watchers: 16
- Forks: 34
- Open Issues: 9
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# AuraSR
![aurasr example](https://storage.googleapis.com/falserverless/gallery/aurasr-animated.webp)GAN-based Super-Resolution for real-world images, a variation of the [GigaGAN](https://mingukkang.github.io/GigaGAN/) paper for image-conditioned upscaling. Torch implementation is based on the unofficial [lucidrains/gigagan-pytorch](https://github.com/lucidrains/gigagan-pytorch) repository.
## Usage
```bash
$ pip install aura-sr
``````python
from aura_sr import AuraSRaura_sr = AuraSR.from_pretrained()
``````python
import requests
from io import BytesIO
from PIL import Imagedef 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(image)
```### Reduce Seam Artifacts
`upscale_4x` upscales the image in tiles that do not overlap. This can result in seams. Use `upscale_4x_overlapped` to reduce seams. This will double the time upscaling by taking an additional pass and averaging the results.