FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing

10 Dec 2024  ·  Yingying Deng, Xiangyu He, Changwang Mei, Peisong Wang, Fan Tang ·

Though Rectified Flows (ReFlows) with distillation offers a promising way for fast sampling, its fast inversion transforms images back to structured noise for recovery and following editing remains unsolved. This paper introduces FireFlow, a simple yet effective zero-shot approach that inherits the startling capacity of ReFlow-based models (such as FLUX) in generation while extending its capabilities to accurate inversion and editing in $8$ steps. We first demonstrate that a carefully designed numerical solver is pivotal for ReFlow inversion, enabling accurate inversion and reconstruction with the precision of a second-order solver while maintaining the practical efficiency of a first-order Euler method. This solver achieves a $3\times$ runtime speedup compared to state-of-the-art ReFlow inversion and editing techniques, while delivering smaller reconstruction errors and superior editing results in a training-free mode. The code is available at $\href{https://github.com/HolmesShuan/FireFlow}{this URL}$.

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Text-based Image Editing PIE-Bench FireFlow CLIPSIM 26.02 # 2
Structure Distance 27.1 # 11
Background PSNR 23.03 # 9
Background LPIPS 123.6 # 13
Text-based Image Editing PIE-Bench FireFlow (Add Q) CLIPSIM 27.33 # 1
Structure Distance 70.9 # 17
Background PSNR 16.49 # 17
Background LPIPS 239.4 # 17

Methods


No methods listed for this paper. Add relevant methods here