SOLA aligns SAM2 object tokens with language to select video mask tracks, achieving state-of-the-art J&F 48.6 on MeViS with 32.9M trainable parameters.
End-to-end referring video object segmentation with mul- timodal 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
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Referring Video Object Segmentation via Language-aligned Track Selection
SOLA aligns SAM2 object tokens with language to select video mask tracks, achieving state-of-the-art J&F 48.6 on MeViS with 32.9M trainable parameters.