PropVG is an end-to-end visual grounding model that generates foreground object proposals with a DETR-style decoder and scores them with sentence- and word-level contrastive learning, achieving state-of-the-art on ten referring expression datasets.
Mask grounding for referring image seg- mentation
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
background 1representative citing papers
citing papers explorer
-
PropVG: End-to-End Proposal-Driven Visual Grounding with Multi-Granularity Discrimination
PropVG is an end-to-end visual grounding model that generates foreground object proposals with a DETR-style decoder and scores them with sentence- and word-level contrastive learning, achieving state-of-the-art on ten referring expression datasets.