ermongroup/ncsnv2

The official PyTorch implementation for NCSNv2 (NeurIPS 2020)

48
/ 100
Emerging

This project helps researchers and artists generate high-resolution, realistic images without needing adversarial training. It takes existing image datasets as input and produces new, diverse image samples that capture the characteristics of the original data. This is ideal for those working on image synthesis, creative AI, or advanced computer vision applications.

321 stars. No commits in the last 6 months.

Use this if you need to generate high-quality images from a dataset that are sharp and diverse, without the complexities often associated with Generative Adversarial Networks (GANs).

Not ideal if your primary goal is exact log-likelihood computation or even higher sample quality, as a subsequent extension (Score-Based Generative Modeling through Stochastic Differential Equations) offers further improvements.

image-generation computer-vision-research creative-ai data-synthesis deep-learning-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

321

Forks

62

Language

Python

License

MIT

Last pushed

Jun 12, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ermongroup/ncsnv2"

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