Introduces CLIP guidance and prior-preservation loss to constrain point-based diffusion image edits within the model's sampling distribution for more consistent results.
Point-based method Point-based image editing was initially introduced by DragGAN[1], which alternates iterative optimization within the latent space
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Drag within Prior Distribution: Text-Conditioned Point-Based Image Editing within Distribution Constraints
Introduces CLIP guidance and prior-preservation loss to constrain point-based diffusion image edits within the model's sampling distribution for more consistent results.