PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator
Abstract
We present Piecewise Rectified Flow (PeRFlow), a flow-based method for accelerating diffusion models. PeRFlow divides the sampling process of generative flows into several time windows and straightens the trajectories in each interval via the reflow operation, thereby approaching piecewise linear flows. PeRFlow achieves superior performance in a few-step generation. Moreover, through dedicated parameterizations, the PeRFlow models inherit knowledge from the pretrained diffusion models. Thus, the training converges fast and the obtained models show advantageous transfer ability, serving as universal plug-and-play accelerators that are compatible with various workflows based on the pre-trained diffusion models.
Cite
Text
Yan et al. "PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator." Neural Information Processing Systems, 2024. doi:10.52202/079017-2497Markdown
[Yan et al. "PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator." Neural Information Processing Systems, 2024.](https://mlanthology.org/neurips/2024/yan2024neurips-perflow/) doi:10.52202/079017-2497BibTeX
@inproceedings{yan2024neurips-perflow,
title = {{PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator}},
author = {Yan, Hanshu and Liu, Xingchao and Pan, Jiachun and Liew, Jun Hao and Liu, Qiang and Feng, Jiashi},
booktitle = {Neural Information Processing Systems},
year = {2024},
doi = {10.52202/079017-2497},
url = {https://mlanthology.org/neurips/2024/yan2024neurips-perflow/}
}