TencentARC/GFPGAN
GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.
Leverages pretrained generative face priors (StyleGAN2) for blind restoration without paired training data, enabling realistic detail recovery even from heavily degraded inputs. Built on PyTorch with BasicSR and facexlib dependencies, featuring modular architecture that supports optional Real-ESRGAN background enhancement and multiple model versions (v1.3, v1.4) optimized for different quality tiers. Includes inference utilities and integrates with Hugging Face Spaces and Replicate for accessible deployment.
37,396 stars and 263,993 monthly downloads. Used by 2 other packages. No commits in the last 6 months. Available on PyPI.
Stars
37,396
Forks
6,283
Language
Python
License
—
Category
Last pushed
Jul 26, 2024
Monthly downloads
263,993
Commits (30d)
0
Dependencies
12
Reverse dependents
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/TencentARC/GFPGAN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
MathiasGruber/PConv-Keras
Unofficial implementation of "Image Inpainting for Irregular Holes Using Partial Convolutions"....
bamos/dcgan-completion.tensorflow
Image Completion with Deep Learning in TensorFlow
leehomyc/Faster-High-Res-Neural-Inpainting
High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis
yu4u/cutout-random-erasing
Cutout / Random Erasing implementation, especially for ImageDataGenerator in Keras
alexandre01/deepsvg
[NeurIPS 2020] Official code for the paper "DeepSVG: A Hierarchical Generative Network for...