analiza aiqbench
This paper proposes a transformer that estimates a camera's 6-DoF pose by matching a single query image to planar primitives from an existing map, rather than relying on costly, multi-stage pipelines and dedicated maps. The authors report state-of-the-art results on dense surface maps and minimal layout models, with code released, though the evidence is limited to the benchmarks they chose and generalization to unseen environments needs independent testing. For AI readers, the notable idea is reusing lightweight, semantically meaningful map structures as the localization target, which could lower map construction and maintenance costs in robotics and AR.
PGGT-Loc: Primitive-Grounded Geometry Transformer for Feed-Forward Camera Localization
arXiv:2610.41833v1 Announce Type: cross Abstract: Camera localization methods are mostly multi-stage pipelines relying on dedicated and complex maps that are costly to construct, store, and maintain over time. Alternatively, well-structured maps that organize scenes with compact yet expressive primitives are often readily available, providing rich structural and semantic cues that humans can interpret. However, their potential for direct camera localization remains largely unexplored due to the significant modality gap between map primitives and visual observations. This motivates a new concept for localization, which estimates camera poses by grounding query observations to map primitives. In this paper, we instantiate the concept in the context of planar primitives with PGGT, the Primitive-Grounded Geometry Transformer network. Given a single query image and a set of planar map primitives, the network is trained to predict multiple localization targets for feed-forward 6-DoF camera localization that generalizes to unseen environments. With this straightforward and scalable architecture, our method pushes the state-of-the-art localization performance on both dense surface maps and minimal layout models. The code and models are available at https://github.com/3dv-casia/PGGT-Loc
Abstract:Camera localization methods are mostly multi-stage pipelines relying on dedicated and complex maps that are costly to construct, store, and maintain over time. Alternatively, well-structured maps that organize scenes with compact yet expressive primitives are often readily available, providing rich structural and semantic cues that humans can interpret. However, their potential for direct camera localization remains largely unexplored due to the significant modality gap between map primitives and visual observations. This motivates a new concept for localization, which estimates camera poses by grounding query observations to map primitives. In this paper, we instantiate the concept in the context of planar primitives with PGGT, the Primitive-Grounded Geometry Transformer network. Given a single query image and a set of planar map primitives, the network is trained to predict multiple localization targets for feed-forward 6-DoF camera localization that generalizes to unseen environments. With this straightforward and scalable architecture, our method pushes the state-of-the-art localization performance on both dense surface maps and minimal layout models. The code and models are available at this https URL
Accepted at NeurIPS 2026. 24 pages, 14 figures, including appendices. Code and models: this https URL
Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
arXiv:2610.41833 [cs.CV]
(or arXiv:2610.41833v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.41833 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 286 cuvinte