A plug-and-play view partitioning scheme using combinatorial graph partitioning on visual dissimilarity and approximated spatial dispersion makes VGGT scalable to large view collections with gains in pose estimation, depth prediction, and reconstruction.
arXiv preprint arXiv:2511.18290 (2025) 2, 4
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Diversity-aware View Partitioning for Scalable VGGT
A plug-and-play view partitioning scheme using combinatorial graph partitioning on visual dissimilarity and approximated spatial dispersion makes VGGT scalable to large view collections with gains in pose estimation, depth prediction, and reconstruction.