不需训练，用最优控制实现精准图像编辑
Training-Free Reward-Guided Image Editing via Trajectory Optimal Control
- 将图像编辑建模为轨迹最优控制问题，动态调整生成路径
- 在多个任务中显著优于现有方法，兼顾奖励提升与原图保真度
- 适合需要快速、无训练图像修改的场景，如设计与内容创作
扩散模型和流匹配模型在高质量图像生成方面取得了显著进展。其中，基于奖励引导的方法在推理过程中可引导生成以满足特定目标。然而，如何将此类方法应用于图像编辑——在保持源图像语义内容的同时增强目标奖励——仍鲜有研究。本文提出一种无需训练的奖励引导图像编辑框架。将编辑过程建模为轨迹最优控制问题：将扩散模型的逆过程视为从源图像出发的可控轨迹，通过迭代更新伴随状态来引导编辑方向。在多种编辑任务上的大量实验表明，该方法显著优于现有的基于反演的无训练引导基线，在不发生奖励滥用的前提下，实现了奖励最大化与源图像保真度之间的优异平衡。
原文摘要 · Abstract (English)
Recent advancements in diffusion and flow-matching models have demonstrated remarkable capabilities in high-fidelity image synthesis. A prominent line of research involves reward-guided guidance, which steers the generation process during inference to align with specific objectives. However, leveraging this reward-guided approach to the task of image editing, which requires preserving the semantic content of the source image while enhancing a target reward, is largely unexplored. In this work, we introduce a novel framework for training-free, reward-guided image editing. We formulate the editing process as a trajectory optimal control problem where the reverse process of a diffusion model is treated as a controllable trajectory originating from the source image, and the adjoint states are iteratively updated to steer the editing process. Through extensive experiments across distinct editing tasks, we demonstrate that our approach significantly outperforms existing inversion-based training-free guidance baselines, achieving a superior balance between reward maximization and fidelity to the source image without reward hacking.
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。