FEC: Three Finetuning-free Methods to Enhance Consistency for Real Image Editing
This work addresses a bottleneck in real image editing for users by enhancing consistency without costly fine-tuning, though it is incremental as it builds on existing diffusion-based methods.
The paper tackled the problem of reconstruction failure in text-conditional image editing, where DDIM Inversion often fails to preserve original image content, by proposing FEC, a set of three finetuning-free sampling methods that ensure successful reconstruction and improve editing performance across various tasks.
Text-conditional image editing is a very useful task that has recently emerged with immeasurable potential. Most current real image editing methods first need to complete the reconstruction of the image, and then editing is carried out by various methods based on the reconstruction. Most methods use DDIM Inversion for reconstruction, however, DDIM Inversion often fails to guarantee reconstruction performance, i.e., it fails to produce results that preserve the original image content. To address the problem of reconstruction failure, we propose FEC, which consists of three sampling methods, each designed for different editing types and settings. Our three methods of FEC achieve two important goals in image editing task: 1) ensuring successful reconstruction, i.e., sampling to get a generated result that preserves the texture and features of the original real image. 2) these sampling methods can be paired with many editing methods and greatly improve the performance of these editing methods to accomplish various editing tasks. In addition, none of our sampling methods require fine-tuning of the diffusion model or time-consuming training on large-scale datasets. Hence the cost of time as well as the use of computer memory and computation can be significantly reduced.