PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator

Hanshu Yan, Xingchao Liu, Jiachun Pan, Jun Hao Liew, qiang liu, Jiashi Feng

Keywords: Flow Model, Diffusion Model, Generative Model, Image Generation

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.

Primary Area: Diffusion based models

Submission Number: 16420