TweezeEdit: Consistent and Efficient Image Editing with Path Regularization
Jianda Mao, Kaibo Wang, Yang Xiang, Kani Chen

TL;DR
TweezeEdit is a novel image editing framework that enhances semantic preservation and efficiency by regularizing the denoising path, enabling real-time edits with minimal steps and improved target alignment.
Contribution
It introduces a tuning- and inversion-free approach that regularizes the entire denoising path, significantly improving semantic retention and editing efficiency over existing methods.
Findings
Outperforms existing methods in semantic preservation and target alignment.
Requires only 12 steps (1.6 seconds) per edit for real-time application.
Effectively maintains source image semantics during editing.
Abstract
Large-scale pre-trained diffusion models empower users to edit images through text guidance. However, existing methods often over-align with target prompts while inadequately preserving source image semantics. Such approaches generate target images explicitly or implicitly from the inversion noise of the source images, termed the inversion anchors. We identify this strategy as suboptimal for semantic preservation and inefficient due to elongated editing paths. We propose TweezeEdit, a tuning- and inversion-free framework for consistent and efficient image editing. Our method addresses these limitations by regularizing the entire denoising path rather than relying solely on the inversion anchors, ensuring source semantic retention and shortening editing paths. Guided by gradient-driven regularization, we efficiently inject target prompt semantics along a direct path using a consistency…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
