UniSpace-3D reports state-of-the-art 3D visual grounding accuracy by mapping point clouds and text into a shared CLIP space and adding contrastive losses and language-guided query selection.
Scanrefer: 3d object local- ization in rgb-d scans using natural language,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
Unified Representation Space for 3D Visual Grounding
UniSpace-3D reports state-of-the-art 3D visual grounding accuracy by mapping point clouds and text into a shared CLIP space and adding contrastive losses and language-guided query selection.