omriav/blended-latent-diffusion
A research implementation of latent diffusion models for localized text-based image editing using masks.
Not currently ranked — collecting fresh signals.
star history
Blended Latent Diffusion enables text-driven local editing of images using latent diffusion models (LDM). The approach combines Blended Diffusion with a text-to-image LDM for faster inference while maintaining editing quality. It addresses image reconstruction limitations inherent in LDMs and handles thin mask scenarios for precise local edits. The method was evaluated against baselines both qualitatively and quantitatively.
Frequently asked
- What is omriav/blended-latent-diffusion?
- A research implementation of latent diffusion models for localized text-based image editing using masks.
- Is blended-latent-diffusion open source?
- Yes — omriav/blended-latent-diffusion is open source, released under the MIT license.
- What language is blended-latent-diffusion written in?
- omriav/blended-latent-diffusion is primarily written in Jupyter Notebook.
- How popular is blended-latent-diffusion?
- omriav/blended-latent-diffusion has 631 stars on GitHub.
- Where can I find blended-latent-diffusion?
- omriav/blended-latent-diffusion is on GitHub at https://github.com/omriav/blended-latent-diffusion.