SL-ASD replaces audio-visual synchronization with face-voice identity matching plus quality-weighted face aggregation, reporting competitive Ego4D validation accuracy with 0.4M learnable parameters, though the comparison uses a modified evaluation protocol.
How to Design a Three-Stage Architecture for Audio-Visual Active Speaker Detection in the Wild,
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.MM 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Face-Voice Association for Audiovisual Active Speaker Detection in Egocentric Recordings
SL-ASD replaces audio-visual synchronization with face-voice identity matching plus quality-weighted face aggregation, reporting competitive Ego4D validation accuracy with 0.4M learnable parameters, though the comparison uses a modified evaluation protocol.