LituRout/RF-Inversion
A rectified flow model for semantic image inversion and editing using stochastic differential equations.
Not currently ranked — collecting fresh signals.
star history
RF-Inversion is a generative AI system that efficiently inverts reference style images and applies semantic edits based on text prompts without requiring textual descriptions of the source images. It uses rectified stochastic differential equations for image generation, enabling applications like content-preserving stylization and controlled image manipulation. The model is available as a Gradio demo, integrated into the Hugging Face diffusers library, and supported in ComfyUI.
Frequently asked
- What is LituRout/RF-Inversion?
- A rectified flow model for semantic image inversion and editing using stochastic differential equations.
- Is RF-Inversion open source?
- Yes — LituRout/RF-Inversion is open source, released under the Apache-2.0 license.
- What language is RF-Inversion written in?
- LituRout/RF-Inversion is primarily written in Python.
- How popular is RF-Inversion?
- LituRout/RF-Inversion has 477 stars on GitHub.
- Where can I find RF-Inversion?
- LituRout/RF-Inversion is on GitHub at https://github.com/LituRout/RF-Inversion.