Edicho: Consistent Image Editing in the Wild
This addresses the problem of maintaining consistency in image edits for users dealing with varied poses, lighting, and environments, though it is incremental as it builds on existing diffusion methods.
The paper tackles the challenge of consistent image editing across diverse real-world images by introducing Edicho, a training-free diffusion-based method that uses explicit image correspondence to guide editing, achieving effective results in various settings.
As a verified need, consistent editing across in-the-wild images remains a technical challenge arising from various unmanageable factors, like object poses, lighting conditions, and photography environments. Edicho steps in with a training-free solution based on diffusion models, featuring a fundamental design principle of using explicit image correspondence to direct editing. Specifically, the key components include an attention manipulation module and a carefully refined classifier-free guidance (CFG) denoising strategy, both of which take into account the pre-estimated correspondence. Such an inference-time algorithm enjoys a plug-and-play nature and is compatible to most diffusion-based editing methods, such as ControlNet and BrushNet. Extensive results demonstrate the efficacy of Edicho in consistent cross-image editing under diverse settings. We will release the code to facilitate future studies.