Metraipolis › archive › dc_bad3bd116aa2
Exact Tile-Based Segmentation Inference for Images Larger than GPU Memory — Europe PMC PMID:39015626
- Source
- Europe PMC (en)
- Original
- https://europepmc.org/article/MED/39015626
- DOI
- 10.6028/jres.126.009
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T04:43:42.106Z
- SHA-256 of the archived text
896dad542ee8700e62a50ec772f73b89fce7e48406816db74af9e4211ab0464e- Stable URL of this page
https://metraipolis.com/documento/dc_bad3bd116aa2
How to cite
Exact Tile-Based Segmentation Inference for Images Larger than GPU Memory — Europe PMC PMID:39015626 — Europe PMC (en), https://europepmc.org/article/MED/39015626 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bad3bd116aa2 (Metraipolis archive dc_bad3bd116aa2, sha256 896dad542ee8700e62a50ec772f73b89fce7e48406816db74af9e4211ab0464e)
BibTeX
@misc{dcbad3bd116aa2,
title = {Exact Tile-Based Segmentation Inference for Images Larger than GPU Memory — Europe PMC PMID:39015626},
howpublished = {Europe PMC (en)},
url = {https://europepmc.org/article/MED/39015626},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bad3bd116aa2; sha256 896dad542ee8700e62a50ec772f73b89fce7e48406816db74af9e4211ab0464e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: Europe PMC (en), https://europepmc.org/article/MED/39015626. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We address the problem of performing exact (tiling-error free) out-of-core semantic segmentation inference of arbitrarily large images using fully convolutional neural networks (FCN). FCN models have the property that once a model is trained, it can be applied on arbitrarily sized images, although it is still constrained by the available GPU memory. This work is motivated by overcoming the GPU memory size constraint without numerically impacting the final result. Our approach is to select a tile size that will fit into GPU memory with a halo border of half the network receptive field. Next, stride across the image by that tile size without the halo. The input tile halos will overlap, while the output tiles join exactly at the seams. Such an approach enables inference to be performed on whole slide microscopy images, such as those generated by a slide scanner. The novelty of this work is in documenting the formulas for determining tile size and stride and then validating them on U-Net and FC-DenseNet architectures. In addition, we quantify the errors due to tiling configurations which do not satisfy the constraints, and we explore the use of architecture effective receptive fields to estimate the tiling parameters. Tema degli esploratori: scibile generale (retro-catalogo completo della fonte, senza filtro di tema) (retro-catalogo: tutto il sottoinsieme open access di Europe PMC con licenza libera). Pubblicazione: Journal of research of the National Institute of Standards and Technology, 2021 (prima pubblicazione 2021-06-03). Identificativi: PMID 39015626 · PMCID PMC10914126 · DOI 10.6028/jres.126.009. Parole chiave degli autori: Artificial intelligence; Convolutional Neural Networks; Effective Receptive Field; Out-of-core Processing; Semantic Segmentation. Pagina: https://europepmc.org/article/MED/39015626 — https://doi.org/10.6028/jres.126.009. Autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza dell'articolo dichiarata da Europe PMC: CC0 (la versione è indicata dall'editore sulla pagina). Abstract riportato senza modifiche al contenuto, con attribuzione; deposito marcato CC0. Letto il 2026-11-04T04:36:26.497Z tramite l'API REST ufficiale di Europe PMC (EMBL-EBI), nel rispetto del robots.txt e del crawl-delay.