Flow Stochastic Segmentation Networks

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ribeiro, Fabio De Sousa, Todd, Omar, Jones, Charles, Kori, Avinash, Mehta, Raghav, Glocker, Ben
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913959277232128
author Ribeiro, Fabio De Sousa
Todd, Omar
Jones, Charles
Kori, Avinash
Mehta, Raghav
Glocker, Ben
author_facet Ribeiro, Fabio De Sousa
Todd, Omar
Jones, Charles
Kori, Avinash
Mehta, Raghav
Glocker, Ben
contents We introduce the Flow Stochastic Segmentation Network (Flow-SSN), a generative segmentation model family featuring discrete-time autoregressive and modern continuous-time flow variants. We prove fundamental limitations of the low-rank parameterisation of previous methods and show that Flow-SSNs can estimate arbitrarily high-rank pixel-wise covariances without assuming the rank or storing the distributional parameters. Flow-SSNs are also more efficient to sample from than standard diffusion-based segmentation models, thanks to most of the model capacity being allocated to learning the base distribution of the flow, constituting an expressive prior. We apply Flow-SSNs to challenging medical imaging benchmarks and achieve state-of-the-art results. Code available: https://github.com/biomedia-mira/flow-ssn.
format Preprint
id arxiv_https___arxiv_org_abs_2507_18838
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Flow Stochastic Segmentation Networks
Ribeiro, Fabio De Sousa
Todd, Omar
Jones, Charles
Kori, Avinash
Mehta, Raghav
Glocker, Ben
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
We introduce the Flow Stochastic Segmentation Network (Flow-SSN), a generative segmentation model family featuring discrete-time autoregressive and modern continuous-time flow variants. We prove fundamental limitations of the low-rank parameterisation of previous methods and show that Flow-SSNs can estimate arbitrarily high-rank pixel-wise covariances without assuming the rank or storing the distributional parameters. Flow-SSNs are also more efficient to sample from than standard diffusion-based segmentation models, thanks to most of the model capacity being allocated to learning the base distribution of the flow, constituting an expressive prior. We apply Flow-SSNs to challenging medical imaging benchmarks and achieve state-of-the-art results. Code available: https://github.com/biomedia-mira/flow-ssn.
title Flow Stochastic Segmentation Networks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.18838