Conceptedit region mask
Region Mask
A binary or soft-weight map with the same dimensions as the input image, indicating where and over what extent an image editing operation should occur. It is used for spatial grounding in decomposed goal-image synthesis.
Timeline (last 90 days)
Oct 20
The decomposed mask method decouples 'where to change' from 'what to change into,' with mask prediction responsible for spatial grounding and the generative model responsible for appearance synthesis, reducing the risk of hallucination or tampering in irrelevant regions.
Unverified50%