TACO uses semantically constrained co-factorization of frozen CLIP/CLAP features, then prompts FC-CLIP with the identified sounding factor, achieving state-of-the-art unsupervised sound-prompted segmentation without training.
Emerging properties in self-supervised vision transformers
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
eess.AS 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
TACO: Training-free Sound Prompted Segmentation via Semantically Constrained Audio-visual CO-factorization
TACO uses semantically constrained co-factorization of frozen CLIP/CLAP features, then prompts FC-CLIP with the identified sounding factor, achieving state-of-the-art unsupervised sound-prompted segmentation without training.