Metraipolis › archive › dc_68bf426195db
G-ray: Ray-Level Relative Geometric Position Encoding in Multi-View Vision Transformers under Camera Heterogeneity — arXiv:2609.15018
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15018
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T01:53:20.746Z
- SHA-256 of the archived text
476ac7e072a988bd8ab45747ec766f4646c2ea0cf1881b56c664f4b25f587016- Stable URL of this page
https://metraipolis.com/documento/dc_68bf426195db
How to cite
G-ray: Ray-Level Relative Geometric Position Encoding in Multi-View Vision Transformers under Camera Heterogeneity — arXiv:2609.15018 — arXiv (en), https://arxiv.org/abs/2609.15018 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_68bf426195db (Metraipolis archive dc_68bf426195db, sha256 476ac7e072a988bd8ab45747ec766f4646c2ea0cf1881b56c664f4b25f587016)
BibTeX
@misc{dc68bf426195db,
title = {G-ray: Ray-Level Relative Geometric Position Encoding in Multi-View Vision Transformers under Camera Heterogeneity — arXiv:2609.15018},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15018},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_68bf426195db; sha256 476ac7e072a988bd8ab45747ec766f4646c2ea0cf1881b56c664f4b25f587016},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15018. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We study relative position encoding for multi-view vision Transformers under camera heterogeneity, including varying fields of view (FoVs) or projection models. Existing rotary relative position encodings commonly use image-plane positional coordinates, producing projection-dependent relative phases and inconsistent geometric cues for cross-projection attention. We introduce G-ray, a ray-level relative position encoding whose rotary phases are parameterized by camera-local ray angles. The same camera-local ray pair induces the same relative phase across projections, providing projection-invariant positional consistency. G-ray can be used directly or integrated with existing encodings, retaining complementary geometric cues without additional learned parameters. We validate G-ray in three host encodings, RoPE, GTA, and RayRoPE, across 3D reconstruction and novel-view synthesis (NVS). Across three heterogeneous 3D reconstruction benchmarks at 50 views, G-ray leads all six averaged metrics and reduces mean pointmap relative error by 45.8% over MapAnything, with calibration supplied to both. Trained exclusively on homogeneous pinhole images, the 3D reconstruction model handles mixed pinhole and non-pinhole inputs without retraining and remains competitive on homogeneous pinhole 3D reconstruction protocols. For NVS, GTA and RayRoPE improve with G-ray under joint viewpoint and FoV variation. The project's webpage is available at this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.15018 — 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 2026-10-16T01:25:23.632Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).