图像编辑中的语义粒度导航
Semantic Granularity Navigation in Image Editing
AI总结：
提出NaviEdit，一种无需训练、推理时控制的解耦方法，通过自一致性约束将编辑进度与模型尺度解耦，在保持结构保真度的同时提升语义可编辑性。
摘要：
尽管扩散模型和流模型具有生成能力，真实图像编辑仍然受到语义可编辑性与结构保真度之间持续权衡的限制。我们将此限制的一个主要原因追溯到现有范式中编辑进度与模型尺度的隐式耦合。在这种耦合下，更强的编辑通常需要访问更嘈杂的状态，这在语义变化被良好定位之前，将计算用于破坏布局。我们引入NaviEdit，一种无需训练的推理时控制器，通过严格的自一致性契约将编辑进度与模型尺度遍历解耦。NaviEdit在rollout级别运行，不改变底层预训练模型。它将尺度视为控制输入，并将固定的步数预算重新分配给语义响应的中间尺度，而不是破坏性的高噪声区域。实验表明，在兼容的编辑器和流骨干网络上，平均增益为正，支持解耦作为一种可移植的推理时控制原则。
英文摘要：
Despite the generative capabilities of diffusion and flow models, real-image editing remains constrained by a persistent trade-off between semantic editability and structural fidelity. We trace a primary cause of this limitation to the implicit coupling of edit progress with model scale in existing paradigms. Under this coupling, stronger edits typically require visiting noisier states, which spends computation on destabilizing layout before the semantic change is well localized. We introduce NaviEdit, a training-free inference-time controller that decouples edit progress from model scale traversal through a strict self-consistency contract. NaviEdit operates at the rollout level and leaves the underlying pretrained model unchanged. It treats scale as a control input and reallocates a fixed step budget toward semantically responsive intermediate scales instead of destructive high-noise regimes. Experiments show positive average gains across compatible editors and flow backbones, supporting decoupling as a portable inference-time control principle.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
