Adams Bashforth Moulton Solver for Inversion and Editing in Rectified Flow
This addresses a bottleneck in downstream applications like reconstruction and editing for users of rectified flow models, though it is incremental as it builds on existing numerical solvers.
The paper tackled the trade-off between fast sampling and high-accuracy solutions in rectified flow models for image and video generation by proposing ABM-Solver, which improved inversion precision and editing quality on multiple high-resolution datasets without extra training.
Rectified flow models have achieved remarkable performance in image and video generation tasks. However, existing numerical solvers face a trade-off between fast sampling and high-accuracy solutions, limiting their effectiveness in downstream applications such as reconstruction and editing. To address this challenge, we propose leveraging the Adams-Bashforth-Moulton (ABM) predictor-corrector method to enhance the accuracy of ODE solving in rectified flow models. Specifically, we introduce ABM-Solver, which integrates a multi step predictor corrector approach to reduce local truncation errors and employs Adaptive Step Size Adjustment to improve sampling speed. Furthermore, to effectively preserve non edited regions while facilitating semantic modifications, we introduce a Mask Guided Feature Injection module. We estimate self-similarity to generate a spatial mask that differentiates preserved regions from those available for editing. Extensive experiments on multiple high-resolution image datasets validate that ABM-Solver significantly improves inversion precision and editing quality, outperforming existing solvers without requiring additional training or optimization.