VipDiff combines optical flow propagation with noise-optimized diffusion sampling to produce temporally coherent and diverse video inpainting results without training.
Learnable gated temporal shift module for deep video inpainting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
VipDiff: Towards Coherent and Diverse Video Inpainting via Training-free Denoising Diffusion Models
VipDiff combines optical flow propagation with noise-optimized diffusion sampling to produce temporally coherent and diverse video inpainting results without training.