无需反演的图像编辑中，通过自适应噪声修正实现结构保真。
SNR-Edit: Structure-Aware Noise Rectification for Inversion-Free Flow-Based Editing
- 基于结构感知噪声修正，动态调整初始噪声以匹配真实图像结构。
- 在SD3和FLUX模型上，像素级指标与VLM评分均显著提升。
- 无需训练或反演，每张图仅增加约1秒计算时间，适合高效编辑场景。
基于流的生成模型实现无需反演的图像编辑，挑战了传统的反演依赖流程。然而，现有方法使用固定高斯噪声构建源轨迹，导致轨迹动态偏倚，引发结构退化或质量下降。为此，我们提出SNR-Edit——一种无需训练的框架，通过自适应噪声控制实现忠实的潜在轨迹修正。其核心机制是利用结构感知噪声修正，将分割约束注入初始噪声，使源轨迹的随机成分锚定于真实图像的隐式反演位置，从而减少源到目标传输过程中的轨迹漂移。该轻量级修改带来更平滑的潜在轨迹，并确保高保真结构保留，无需模型调优或反演。在PIE-Bench和SNR-Bench上的评估表明，SNR-Edit在像素级指标和基于视觉语言模型的评分上表现优异，且每张图像仅增加约1秒开销。
原文摘要 · Abstract (English)
Inversion-free image editing using flow-based generative models challenges the prevailing inversion-based pipelines. However, existing approaches rely on fixed Gaussian noise to construct the source trajectory, leading to biased trajectory dynamics and causing structural degradation or quality loss. To address this, we introduce SNR-Edit, a training-free framework achieving faithful Latent Trajectory Correction via adaptive noise control. Mechanistically, SNR-Edit uses structure-aware noise rectification to inject segmentation constraints into the initial noise, anchoring the stochastic component of the source trajectory to the real image's implicit inversion position and reducing trajectory drift during source--target transport. This lightweight modification yields smoother latent trajectories and ensures high-fidelity structural preservation without requiring model tuning or inversion. Across SD3 and FLUX, evaluations on PIE-Bench and SNR-Bench show that SNR-Edit delivers performance on pixel-level metrics and VLM-based scoring, while adding only about 1s overhead per image.
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。