Metraipolis › archive › dc_875cfebc68f3
StreamRig: Exploiting Intra-Rig Geometry for Streaming Multi-Camera Odometry — arXiv:2609.40244
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.40244
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T14:46:39.056Z
- SHA-256 of the archived text
b30eaebb06d25164fc282e36bcb238379578b0b0c15e992a679310978da9b344- Stable URL of this page
https://metraipolis.com/documento/dc_875cfebc68f3
How to cite
StreamRig: Exploiting Intra-Rig Geometry for Streaming Multi-Camera Odometry — arXiv:2609.40244 — arXiv (en), https://arxiv.org/abs/2609.40244 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_875cfebc68f3 (Metraipolis archive dc_875cfebc68f3, sha256 b30eaebb06d25164fc282e36bcb238379578b0b0c15e992a679310978da9b344)
BibTeX
@misc{dc875cfebc68f3,
title = {StreamRig: Exploiting Intra-Rig Geometry for Streaming Multi-Camera Odometry — arXiv:2609.40244},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.40244},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_875cfebc68f3; sha256 b30eaebb06d25164fc282e36bcb238379578b0b0c15e992a679310978da9b344},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40244. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Mobile robots and vehicles carry synchronized multi-camera rigs, yet many streaming 3D foundation models are designed for monocular input, leaving efficient use of rig geometry a challenge. We present StreamRig, a freeze-and-stream framework that builds causal streaming odometry for calibrated rigs on a frozen multi-view 3D foundation model. The frozen front-end jointly perceives the synchronized views using rig calibration. A Rig-Resampler compresses their features, a CausalBridge applies causal attention with a key-value cache, and a lightweight head regresses rig poses. A periodic re-anchoring protocol supports stable pose estimation over long sequences. Only these modules are trained, 74.6M parameters in total, with relative poses as the sole supervision. Our two-stage training strategy combines group relocalization pretraining with causal rig training to transfer the geometric priors of the frozen front-end and the alignment ability of the pretrained modules to streaming odometry. We evaluate on NCLT, TartanGround, KITTI-360, and our self-collected humanoid-robot dataset ZJH, where training uses only simulation and real-world evaluation is zero-shot. Across all four datasets, StreamRig achieves lower translation and rotation drift than the evaluated non-oracle monocular streaming and rig-aware offline models, while maintaining low inference cost. Ablations and controlled camera-count experiments identify the sources of these gains. We further examine how longer training windows affect inference over longer horizons. Code has been released at this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.40244 — 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 2026-10-07T14:19:00.855Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).