As the project describes, it's an inversion-free editing (InfEdit) method that allows for consistent editing at both the semantic and spatial levels, catering to intricate modifications without compromising on the image's integrity and explicit inversion. Through extensive experiments, InfEdit shows strong performance in complex editing tasks and also maintains a seamless workflow (less than 3 seconds on one A40), demonstrating the potential for real-time applications.
Paper: https://sihanxu.github.io/InfEdit/docs/infedit.pdf InfEdit on ArXiv: Not yet published
GitHub Repo: https://github.com/sled-group/InfEdit/tree/website
source:
source prompt: A painting of a waterfall InfEdit prompt: [+and angels] in the mountains
InfEdited result:
Please visit the Infedit demo website: http://sled-snowbird.eecs.umich.edu/
Project Home: https://sled-group.github.io/InfEdit/