VinAIResearch/WaveDiff
A research implementation of wavelet-based diffusion models that accelerate image generation by leveraging frequency-domain decomposition.
Not currently ranked — collecting fresh signals.
star history
WaveDiff is a wavelet-based diffusion scheme for image generation that decomposes images into low-and-high frequency components using wavelet transforms at both image and feature levels. The approach adaptively accelerates the sampling process while maintaining generation quality. The implementation includes training and evaluation code for datasets including CelebA-HQ, CIFAR-10, LSUN-Church, and STL-10, targeting state-of-the-art inference speed for diffusion models.
Frequently asked
- What is VinAIResearch/WaveDiff?
- A research implementation of wavelet-based diffusion models that accelerate image generation by leveraging frequency-domain decomposition.
- Is WaveDiff open source?
- Yes — VinAIResearch/WaveDiff is open source, released under the AGPL-3.0 license.
- What language is WaveDiff written in?
- VinAIResearch/WaveDiff is primarily written in Python.
- How popular is WaveDiff?
- VinAIResearch/WaveDiff has 440 stars on GitHub.
- Where can I find WaveDiff?
- VinAIResearch/WaveDiff is on GitHub at https://github.com/VinAIResearch/WaveDiff.