SAB3R unifies 3D reconstruction and open-vocabulary segmentation in a single feed-forward network trained by distilling CLIP and DINOv2 features into MASt3R.
Unsupervised scale-consistent depth learning from video.In- ternational Journal of Computer Vision (IJCV), 2021
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
unclear 1representative citing papers
citing papers explorer
-
SAB3R: Semantic-Augmented Backbone in 3D Reconstruction
SAB3R unifies 3D reconstruction and open-vocabulary segmentation in a single feed-forward network trained by distilling CLIP and DINOv2 features into MASt3R.