通过动态融合隐变量实现语义一致的图像编辑，轻量高效且无需修改模型。
LatentEdit: Adaptive Latent Control for Consistent Semantic Editing
- 动态融合当前与参考隐变量，精准控制编辑区域。
- 在8-15步内超越当前最佳方法，保持高保真与可编辑性平衡。
- 无需模型修改，适配UNet和DiT架构，适合实时应用。
基于扩散模型的图像编辑近年来取得显著进展，但如何在保持背景相似性的同时实现高质量编辑，且不牺牲速度或内存效率，仍是挑战。本文提出LatentEdit，一种自适应隐变量融合框架，动态结合当前隐变量与从源图反演得到的参考隐变量。通过在高相似度、语义重要区域保留源特征，同时在其他区域根据目标提示生成新内容，实现细粒度、可控编辑。该方法无需内部模型修改或复杂注意力机制，为轻量级、即插即用方案，兼容基于UNet和DiT的架构。在PIE-Bench数据集上的大量实验表明，LatentEdit在保真度与可编辑性之间达到最优平衡，即使在8-15步内也优于现有最先进方法。其无反演变体进一步将神经函数评估次数减半，并消除中间变量存储需求，显著提升实时部署效率。
原文摘要 · Abstract (English)
Diffusion-based Image Editing has achieved significant success in recent years. However, it remains challenging to achieve high-quality image editing while maintaining the background similarity without sacrificing speed or memory efficiency. In this work, we introduce LatentEdit, an adaptive latent fusion framework that dynamically combines the current latent code with a reference latent code inverted from the source image. By selectively preserving source features in high-similarity, semantically important regions while generating target content in other regions guided by the target prompt, LatentEdit enables fine-grained, controllable editing. Critically, the method requires no internal model modifications or complex attention mechanisms, offering a lightweight, plug-and-play solution compatible with both UNet-based and DiT-based architectures. Extensive experiments on the PIE-Bench dataset demonstrate that our proposed LatentEdit achieves an optimal balance between fidelity and editability, outperforming the state-of-the-art method even in 8-15 steps. Additionally, its inversion-free variant further halves the number of neural function evaluations and eliminates the need for storing any intermediate variables, substantially enhancing real-time deployment efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。