Metraipolis › archive › dc_4382ca94a4a2

DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers — arXiv:2609.12557

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.12557
License (as archived)
CC0-1.0
Archived
2026-10-16T03:57:39.674Z
SHA-256 of the archived text
7150328242839fc3f1d10baac0bb07e82c9bf6857e71b25f7cb81b252f21b512
Stable URL of this page
https://metraipolis.com/documento/dc_4382ca94a4a2

How to cite

DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers — arXiv:2609.12557 — arXiv (en), https://arxiv.org/abs/2609.12557 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4382ca94a4a2 (Metraipolis archive dc_4382ca94a4a2, sha256 7150328242839fc3f1d10baac0bb07e82c9bf6857e71b25f7cb81b252f21b512)
BibTeX
@misc{dc4382ca94a4a2,
  title = {DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers — arXiv:2609.12557},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.12557},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4382ca94a4a2; sha256 7150328242839fc3f1d10baac0bb07e82c9bf6857e71b25f7cb81b252f21b512},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.12557. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present DRS-VPT, a feed-forward transformer architecture for foundational image-to-scan registration. Given query images and a reference 3D point cloud, the model predicts the scan pose and point map alongside the poses and point maps of each camera, all expressed in the first camera's frame. It additionally predicts a coarse-to- fine pyramid of per-point and per-pixel features for direct reprojective alignment of the scan to the first image. This formulation unifies downstream tasks such as camera-LiDAR calibration in autonomous driving and indoor camera-to-map relocalization. A single DRS-VPT model achieves state-of-the-art performance for image-to-LiDAR registration in autonomous driving, competitive indoor relocalization without training map-specific weights, and strong zero-shot transfer to unseen environments. We also show qualitatively that the model learns complex scan-to-image projection properties such as occlusion of back-facing points.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-14 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.12557 — 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 2026-10-16T03:44:26.274Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).