Metraipolis › archive › dc_3bf43becf8a5

Hybrid supervised instance segmentation by learning label noise suppression — arXiv:2610.20412

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.20412
License (as archived): CC0-1.0
Archived: 2026-10-15T21:57:18.863Z
SHA-256 of the archived text: aecb182f151f81067489d00bb3ddebbb08382dbf5c4f05b477ddd7e9c4254872
Stable URL of this page: https://metraipolis.com/documento/dc_3bf43becf8a5

How to cite

Hybrid supervised instance segmentation by learning label noise suppression — arXiv:2610.20412 — arXiv (en), https://arxiv.org/abs/2610.20412 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3bf43becf8a5 (Metraipolis archive dc_3bf43becf8a5, sha256 aecb182f151f81067489d00bb3ddebbb08382dbf5c4f05b477ddd7e9c4254872)

BibTeX

@misc{dc3bf43becf8a5,
  title = {Hybrid supervised instance segmentation by learning label noise suppression — arXiv:2610.20412},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20412},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3bf43becf8a5; sha256 aecb182f151f81067489d00bb3ddebbb08382dbf5c4f05b477ddd7e9c4254872},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20412. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

To reach top accuracy, current fully supervised instance segmentation methods severely rely on large-scale pixel-wise labeled datasets. They are usually expensive and time-consuming to obtain. Though weakly or semi-supervised methods utilize cheap bounding box labeled, image-level labeled or unlabeled samples to save the labeling cost, their performance is largely sacrificed. To save labeling cost without losing much performance, in this paper, we present a pipeline that can utilize economical bounding box labels and accurate pixel-wise labels in a hybrid way. Specifically, we design two ancillary models to learn label noise suppression and obtain accurate pseudo pixel-wise labels from bounding box labels for training. One is designed to suppress mislabeling between foreground and background, and the other is designed to suppress noise from mislabeling of instances. Moreover, we exploit category-aware spatial attention module, category constraint module, instance constraint module, and self-learning training approach to improve the accuracy of pseudo labels. Experiments on the PASCAL VOC 2012 and the Cityscapes datasets show that our method can achieve competitive performance with much less labeling cost.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.20412 — 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 2026-10-15T21:38:29.539Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).