VinAIResearch/SA-DPM

Official PyTorch implementation of "On Inference Stability for Diffusion Models" (AAAI'24)

28
/ 100
Experimental

This project helps researchers and practitioners in generative AI improve the quality of images created by Denoising Probabilistic Models (DPMs). It takes existing DPM training data and models as input and applies a novel 'sequence-aware' loss function during training. The output is a more stable and higher-quality image generation model, suitable for those working with advanced image synthesis.

No commits in the last 6 months.

Use this if you are developing or training diffusion models and want to enhance the stability and visual quality of the generated images, especially for complex datasets like human faces or animal images.

Not ideal if you are looking for an out-of-the-box image generation tool or if you do not have experience with training deep learning models and managing datasets like CelebA or FFHQ.

generative-ai image-synthesis diffusion-models deep-learning-research computer-vision
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 4 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

Python

License

AGPL-3.0

Last pushed

Jul 23, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/VinAIResearch/SA-DPM"

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