Metraipolis › archive › dc_de50983cbf3d

FUSEye: Training-Light Fisheye Detection with Overlapping Views and Zero-Initialized Adapters — arXiv:2610.02799

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.02799
License (as archived)
CC0-1.0
Archived
2026-10-06T09:07:14.262Z
SHA-256 of the archived text
b62824ea82596407f4cfa332e6a2ce59e499c3aabfbeb6e4d37a12033e473753
Stable URL of this page
https://metraipolis.com/documento/dc_de50983cbf3d

How to cite

FUSEye: Training-Light Fisheye Detection with Overlapping Views and Zero-Initialized Adapters — arXiv:2610.02799 — arXiv (en), https://arxiv.org/abs/2610.02799 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_de50983cbf3d (Metraipolis archive dc_de50983cbf3d, sha256 b62824ea82596407f4cfa332e6a2ce59e499c3aabfbeb6e4d37a12033e473753)
BibTeX
@misc{dcde50983cbf3d,
  title = {FUSEye: Training-Light Fisheye Detection with Overlapping Views and Zero-Initialized Adapters — arXiv:2610.02799},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.02799},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_de50983cbf3d; sha256 b62824ea82596407f4cfa332e6a2ce59e499c3aabfbeb6e4d37a12033e473753},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02799. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fisheye cameras give mobile robots a single-sensor, low-cost view of their surroundings, yet the COCO-pretrained detectors that practitioners routinely reuse fail on them: strong radial distortion warps local image structure, while boundary compression shrinks objects to near-invisible sizes. Full fine-tuning closes much of the gap but requires abundant fisheye labels and compute. We present FUSEye, a training-light framework that turns a frozen-backbone COCO-pretrained extra-large YOLO26 detector (YOLO26-x) into a fisheye detector. FUSEye adds roughly 227k new parameters while updating the inserted modules and the pretrained detection head. It addresses the transfer gap at three causally linked levels. At the input level, overlapping grid view generation and box remapping (GridViews) enlarge compressed boundary regions. At the feature level, zero-initialized residual adapters (Z-Adapters) correct distortion-induced feature misalignment. At the decision level, learned cross-projection agreement fusion (AgreeFusion) promotes low-confidence detections only when they are supported by consistent evidence across multiple views. On the WoodScape surround-view fisheye benchmark, FUSEye raises YOLO26-x from 0.148 to 0.266 mAP50 and retains 84.3% fully fine-tuned accuracy. Moreover, randomly using only 25% of the labeled training images, FUSEye achieves 0.2597 mAP50, retaining 97.6% of its full-label performance. FUSEye also consistently improves YOLOv8-11 detectors, showing that the recipe is architecture-agnostic. Source code will be available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO); Image and Video Processing (eess.IV).
Pagina del lavoro: https://arxiv.org/abs/2610.02799 — 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 2026-10-06T08:21:44.173Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).