BVINet performs blind video inpainting by jointly predicting corruption masks and completing frames, removing the need for mask annotations at inference.
Restoration of analog videos using swin-unet
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
-
BVINet: Unlocking Blind Video Inpainting with Zero Annotations
BVINet performs blind video inpainting by jointly predicting corruption masks and completing frames, removing the need for mask annotations at inference.