KV-Edit: Training-Free Image Editing for Precise Background Preservation

TL;DR

KV-Edit is a training-free image editing method that preserves background consistency by using KV cache in DiTs, avoiding complex training and improving quality in seamless content integration.

Contribution

It introduces a novel training-free approach utilizing KV cache in DiTs for background preservation, with optimized space complexity and broad compatibility.

Findings

Outperforms existing methods in background and image quality

Compatible with any DiT-based model without additional training

Achieves $O(1)$ space complexity with inversion-free optimization

Abstract

Background consistency remains a significant challenge in image editing tasks. Despite extensive developments, existing works still face a trade-off between maintaining similarity to the original image and generating content that aligns with the target. Here, we propose KV-Edit, a training-free approach that uses KV cache in DiTs to maintain background consistency, where background tokens are preserved rather than regenerated, eliminating the need for complex mechanisms or expensive training, ultimately generating new content that seamlessly integrates with the background within user-provided regions. We further explore the memory consumption of the KV cache during editing and optimize the space complexity to $O(1)$ using an inversion-free method. Our approach is compatible with any DiT-based generative model without additional training. Experiments demonstrate that KV-Edit significantly outperforms existing approaches in terms of both background and image quality, even surpassing training-based methods. Project webpage is available at https://xilluill.github.io/projectpages/KV-Edit