PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913487669690368 |
|---|---|
| author | Yan, Hanshu Liu, Xingchao Pan, Jiachun Liew, Jun Hao Liu, Qiang Feng, Jiashi |
| author_facet | Yan, Hanshu Liu, Xingchao Pan, Jiachun Liew, Jun Hao Liu, Qiang Feng, Jiashi |
| contents | 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. Codes for training and inference are publicly released. https://github.com/magic-research/piecewise-rectified-flow |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_07510 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator Yan, Hanshu Liu, Xingchao Pan, Jiachun Liew, Jun Hao Liu, Qiang Feng, Jiashi Machine Learning 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. Codes for training and inference are publicly released. https://github.com/magic-research/piecewise-rectified-flow |
| title | PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2405.07510 |