Metraipolis › archive › dc_5cea4ea3102e

RelationVGGT: Visual Geometry Transformers for 3D Spatial Relation Segmentation — arXiv:2610.00970

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.00970
License (as archived): CC0-1.0
Archived: 2026-10-06T12:12:11.410Z
SHA-256 of the archived text: 8c4623af4cb8db2757113828f44650b625201a5feaa1e6bbe572bd1759509948
Stable URL of this page: https://metraipolis.com/documento/dc_5cea4ea3102e

How to cite

RelationVGGT: Visual Geometry Transformers for 3D Spatial Relation Segmentation — arXiv:2610.00970 — arXiv (en), https://arxiv.org/abs/2610.00970 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5cea4ea3102e (Metraipolis archive dc_5cea4ea3102e, sha256 8c4623af4cb8db2757113828f44650b625201a5feaa1e6bbe572bd1759509948)

BibTeX

@misc{dc5cea4ea3102e,
  title = {RelationVGGT: Visual Geometry Transformers for 3D Spatial Relation Segmentation — arXiv:2610.00970},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.00970},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5cea4ea3102e; sha256 8c4623af4cb8db2757113828f44650b625201a5feaa1e6bbe572bd1759509948},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00970. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent advances in 3D reconstruction have progressed from per-scene optimization to feed-forward inference, and semantic scene understanding has followed suit -- yet existing methods remain confined to object-centric perception, neglecting spatial relations between objects. We formulate 3D spatial relation segmentation in a feed-forward, pose-free multi-view setting: given a visually specified subject and a relational text query, the model segments the target across views without receiving its category name. To this end, we propose RelationVGGT, a novel feed-forward framework that integrates semantic features from a visual foundation model with geometry-aware representations from a 3D geometry foundation model and leverages a relation transformer for subject-conditioned, cross-view relation prediction -- requiring neither per-scene optimization nor known camera poses. We additionally provide a fully automated annotation pipeline built on ScanNet++ with VLMs and LLMs, enabling scalable training data generation for this new task.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.00970 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T10:21:28.139Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).