STR-Match edits videos without retraining by matching source and target 'spatiotemporal relevance scores' derived from attention maps during latent optimization.
Diffusion-based conditional image editing through optimized inference with guidance
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
STR-Match: Matching SpatioTemporal Relevance Score for Training-Free Video Editing
STR-Match edits videos without retraining by matching source and target 'spatiotemporal relevance scores' derived from attention maps during latent optimization.