Metraipolis › archive › dc_c0fa4371bbfe

DualPathOcc: Dual-Resolution BEV Encoder for 3D Occupancy Prediction — arXiv:2609.06370

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.06370
License (as archived): CC0-1.0
Archived: 2026-10-16T15:06:59.963Z
SHA-256 of the archived text: f203766699245ac89d9bc6e360eead10b0bfbf28d9fbd864b607d3199b516b84
Stable URL of this page: https://metraipolis.com/documento/dc_c0fa4371bbfe

How to cite

DualPathOcc: Dual-Resolution BEV Encoder for 3D Occupancy Prediction — arXiv:2609.06370 — arXiv (en), https://arxiv.org/abs/2609.06370 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c0fa4371bbfe (Metraipolis archive dc_c0fa4371bbfe, sha256 f203766699245ac89d9bc6e360eead10b0bfbf28d9fbd864b607d3199b516b84)

BibTeX

@misc{dcc0fa4371bbfe,
  title = {DualPathOcc: Dual-Resolution BEV Encoder for 3D Occupancy Prediction — arXiv:2609.06370},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06370},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c0fa4371bbfe; sha256 f203766699245ac89d9bc6e360eead10b0bfbf28d9fbd864b607d3199b516b84},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06370. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Predicting 3D occupancy from multi-view images requires preserving geometric detail during 2D-to-3D lifting while reasoning over sparse, volumetric scene representations. We present DualPathOcc, a camera-based framework that combines a Spatial Enhancer for high-resolution feature aggregation before BEV compression, a SENet-augmented dual-path BEV encoder for local-global context modeling, and height-aware weighted cross-entropy for near-ground occupancy. The final model is optimized with occupancy supervision and no explicit depth loss. On single-frame Occ3D-nuScenes, DualPathOcc achieves 37.37 mIoU. We further analyze how surface-centered depth targets interact with volumetric occupancy learning.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.06370 — 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 2026-10-16T14:49:57.079Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).