AdaEdit通过自适应调节时间与通道，解决图像编辑中背景保留与内容生成的矛盾。
AdaEdit: Adaptive Temporal and Channel Modulation for Flow-Based Image Editing
- 采用渐进式注入策略，用连续衰减函数替代硬开关，平滑过渡源特征与目标特征
- 基于通道重要性差异施加差异化扰动，提升编辑相关通道响应，保护结构通道
- 在PIE-Bench上性能超越基线，保持高语义相似度，且兼容多种求解器
基于反演的流匹配模型图像编辑已成为无需训练、文本引导的图像操作范式。该范式的核心挑战是注入困境：在去噪过程中注入源特征可保留原始图像背景，但会抑制模型生成编辑内容的能力。现有方法采用固定注入策略——二值化时间调度、均匀空间混合比、不区分通道的潜在扰动——忽略了注入需求在时间与通道维度上的异质性。本文提出AdaEdit，一种无需训练的自适应编辑框架，通过两项互补创新解决该问题。首先，提出渐进式注入策略，以Sigmoid、余弦或线性衰减函数取代硬性二值截断，实现从源特征保留到目标特征生成的平滑过渡，消除特征不连续伪影。其次，引入通道选择性潜在扰动，基于反演潜变量与随机潜变量的分布差距估计各通道重要性，并相应施加差异化的扰动强度——强扰动编辑相关通道，保护结构编码通道。在PIE-Bench基准（700张图像，10类编辑）上的大量实验表明，与强基线相比，AdaEdit在LPIPS上降低8.7%，SSIM提升2.6%，PSNR提升2.3%，同时保持良好的CLIP相似度。AdaEdit完全即插即用，兼容Euler、RF-Solver和FireFlow等多种ODE求解器。代码已开源。
原文摘要 · Abstract (English)
Inversion-based image editing in flow matching models has emerged as a powerful paradigm for training-free, text-guided image manipulation. A central challenge in this paradigm is the injection dilemma: injecting source features during denoising preserves the background of the original image but simultaneously suppresses the model's ability to synthesize edited content. Existing methods address this with fixed injection strategies -- binary on/off temporal schedules, uniform spatial mixing ratios, and channel-agnostic latent perturbation -- that ignore the inherently heterogeneous nature of injection demand across both the temporal and channel dimensions. In this paper, we present AdaEdit, a training-free adaptive editing framework that resolves this dilemma through two complementary innovations. First, we propose a Progressive Injection Schedule that replaces hard binary cutoffs with continuous decay functions (sigmoid, cosine, or linear), enabling a smooth transition from source-feature preservation to target-feature generation and eliminating feature discontinuity artifacts. Second, we introduce Channel-Selective Latent Perturbation, which estimates per-channel importance based on the distributional gap between the inverted and random latents and applies differentiated perturbation strengths accordingly -- strongly perturbing edit-relevant channels while preserving structure-encoding channels. Extensive experiments on the PIE-Bench benchmark (700 images, 10 editing types) demonstrate that AdaEdit achieves an 8.7% reduction in LPIPS, a 2.6% improvement in SSIM, and a 2.3% improvement in PSNR over strong baselines, while maintaining competitive CLIP similarity. AdaEdit is fully plug-and-play and compatible with multiple ODE solvers including Euler, RF-Solver, and FireFlow. Code is available at https://github.com/leeguandong/AdaEdit
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。