Metraipolis › archive › dc_b6aa717ea8bf
Evaluating SAM-based and Convolutional Architectures in Thin Retinal Vessel Segmentation: Model Capacity, Computational Cost and Foundation Pre-training — arXiv:2611.00396
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00396
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:40:38.105Z
- SHA-256 of the archived text
090eb0979d0c57dffcf86dc8855d7b6d7f3af1f1e41875a7de21100addcd2800- Stable URL of this page
https://metraipolis.com/documento/dc_b6aa717ea8bf
How to cite
Evaluating SAM-based and Convolutional Architectures in Thin Retinal Vessel Segmentation: Model Capacity, Computational Cost and Foundation Pre-training — arXiv:2611.00396 — arXiv (en), https://arxiv.org/abs/2611.00396 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b6aa717ea8bf (Metraipolis archive dc_b6aa717ea8bf, sha256 090eb0979d0c57dffcf86dc8855d7b6d7f3af1f1e41875a7de21100addcd2800)
BibTeX
@misc{dcb6aa717ea8bf,
title = {Evaluating SAM-based and Convolutional Architectures in Thin Retinal Vessel Segmentation: Model Capacity, Computational Cost and Foundation Pre-training — arXiv:2611.00396},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00396},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b6aa717ea8bf; sha256 090eb0979d0c57dffcf86dc8855d7b6d7f3af1f1e41875a7de21100addcd2800},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00396. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In retinal vessel segmentation, plain Vision Transformers (ViTs) are commonly expected to perform worse than convolutional neural networks (CNNs) on thin vessels, for reasons such as their coarse patch tokenization, their weak locality bias and their need for large training sets. Although this expectation is widely shared in the literature, how plain ViTs actually compare with CNNs on thin vessels has not been systematically quantified. We address this question by comparing a model with a plain ViT encoder (MedSAM), a hybrid Transformer (LiteMedSAM) and three CNNs of different sizes on three retinal datasets (DRIVE, HRF, FIVES) and, as a second-domain check, a microscopy dataset (VessMap). We measured the centerline recall and precision of thin vessels and compared the models while also considering their computational cost and how many epochs they needed to train. We found that the pre-trained MedSAM had a centerline recall of thin vessels that was not significantly different from that of the best CNN on all four datasets (-0.6 to +2.5 percentage points). A compact plain ViT obtained through architecture search but trained from scratch did not retain this performance. For pre-trained plain ViTs, our results do not support the expectation that they miss thin vessels, although similar thin-vessel performance can be obtained with much smaller models. 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/2611.00396 — 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 2026-11-03T11:06:37.419Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).