TweezeEdit: Consistent and Efficient Image Editing with Path Regularization
This addresses the challenge of efficient and consistent image editing for users of diffusion models, though it appears incremental as it builds on existing methods with improvements in speed and preservation.
The paper tackles the problem of over-alignment and poor semantic preservation in text-guided image editing with diffusion models by proposing TweezeEdit, a tuning- and inversion-free framework that regularizes the denoising path, resulting in superior performance and requiring only 12 steps (1.6 seconds per edit).
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 model. Extensive experiments demonstrate TweezeEdit's superior performance in semantic preservation and target alignment, outperforming existing methods. Remarkably, it requires only 12 steps (1.6 seconds per edit), underscoring its potential for real-time applications.