nupurkmr9/vision-aided-gan

Ensembling Off-the-shelf Models for GAN Training (CVPR 2022 Oral)

40
/ 100
Emerging

This project helps researchers and developers working with Generative Adversarial Networks (GANs) to create more realistic and high-quality synthetic images. It takes existing datasets and leverages the collective 'knowledge' of a bank of pre-trained computer vision models to guide the GAN's training, producing outputs that closely resemble real-world images. This is ideal for those who need to generate highly convincing visual content, especially when real data is scarce.

421 stars. No commits in the last 6 months.

Use this if you are developing image generation models and want to significantly improve the realism and quality of your synthetic images, even with limited training data.

Not ideal if your primary goal is not image generation, or if you need to train models from scratch without leveraging existing vision model knowledge.

generative-ai synthetic-media-creation computer-vision-research image-synthesis machine-learning-development
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

421

Forks

32

Language

Python

License

MIT

Category

gan-based-t2i

Last pushed

Sep 09, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/nupurkmr9/vision-aided-gan"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.