Spatially conditioning a diffusion model by concatenating a reference human image with the noisy target, and adding causal self-attention, improves appearance consistency in human image and video animation.
Masactrl: Tuning-free mutual self-attention control for consistent image synthesis and editing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Consistent Human Image and Video Generation with Spatially Conditioned Diffusion
Spatially conditioning a diffusion model by concatenating a reference human image with the noisy target, and adding causal self-attention, improves appearance consistency in human image and video animation.