Metraipolis › archive › dc_ddbebb3916d2
GeoUniPR: A Geometry-Consistent Unified Framework for Cross-Modal Place Recognition — arXiv:2608.11263
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.11263
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T05:49:18.078Z
- SHA-256 of the archived text
2acf3f596385323f6d92da6c343815f6a79f0c59c228d19536541d22fdb412ca- Stable URL of this page
https://metraipolis.com/documento/dc_ddbebb3916d2
How to cite
GeoUniPR: A Geometry-Consistent Unified Framework for Cross-Modal Place Recognition — arXiv:2608.11263 — arXiv (en), https://arxiv.org/abs/2608.11263 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ddbebb3916d2 (Metraipolis archive dc_ddbebb3916d2, sha256 2acf3f596385323f6d92da6c343815f6a79f0c59c228d19536541d22fdb412ca)
BibTeX
@misc{dcddbebb3916d2,
title = {GeoUniPR: A Geometry-Consistent Unified Framework for Cross-Modal Place Recognition — arXiv:2608.11263},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.11263},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ddbebb3916d2; sha256 2acf3f596385323f6d92da6c343815f6a79f0c59c228d19536541d22fdb412ca},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11263. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Cross-modal place recognition (CMPR) aims to identify the same location across heterogeneous sensing modalities, such as vision and LiDAR. Existing methods commonly bridge the modality gap using complex alignment modules, multi-stage training, or full fine-tuning of pretrained backbones. In this work, we revisit CMPR from the perspective of geometric consistency and propose GeoUniPR, a unified and concise geometry-consistent framework. GeoUniPR reduces cross-modal discrepancy at the representation level by projecting LiDAR point clouds into the camera perspective to construct Geometry-Consistent depth image views (DIV), which establish direct RGB-LiDAR correspondence. We further augment DIV with native LiDAR cues, including intensity and surface-normal information, yielding a multi-channel geometric representation that improves structural consistency. Based on this representation, GeoUniPR learns a unified embedding space using two modality-specific ViT-based encoders with identical architectures, trained through parameter-efficient adaptation without auxiliary alignment modules, multi-stage training, or full backbone fine-tuning. In addition, we introduce Spatially-Consistent InfoNCE (SC-InfoNCE), a CMPR-specific contrastive objective that suppresses distance-induced false negatives under spatial continuity. Extensive experiments on KITTI and KITTI-360 demonstrate that GeoUniPR achieves state-of-the-art (SOTA) performance in both same-modal and cross-modal place recognition, with strong cross-dataset generalization. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-13 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.11263 — 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 2026-10-19T05:42:25.856Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).