Maybe you don't need a U-Net: convolutional feature upsampling for materials micrograph segmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Docherty, Ronan, Vamvakeros, Antonis, Cooper, Samuel J.
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912558949072896
author Docherty, Ronan
Vamvakeros, Antonis
Cooper, Samuel J.
author_facet Docherty, Ronan
Vamvakeros, Antonis
Cooper, Samuel J.
contents Feature foundation models - usually vision transformers - offer rich semantic descriptors of images, useful for downstream tasks such as (interactive) segmentation and object detection. For computational efficiency these descriptors are often patch-based, and so struggle to represent the fine features often present in micrographs; they also struggle with the large image sizes present in materials and biological image analysis. In this work, we train a convolutional neural network to upsample low-resolution (i.e, large patch size) foundation model features with reference to the input image. We apply this upsampler network (without any further training) to efficiently featurise and then segment a variety of microscopy images, including plant cells, a lithium-ion battery cathode and organic crystals. The richness of these upsampled features admits separation of hard to segment phases, like hairline cracks. We demonstrate that interactive segmentation with these deep features produces high-quality segmentations far faster and with far fewer labels than training or finetuning a more traditional convolutional network.
format Preprint
id arxiv_https___arxiv_org_abs_2508_21529
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Maybe you don't need a U-Net: convolutional feature upsampling for materials micrograph segmentation
Docherty, Ronan
Vamvakeros, Antonis
Cooper, Samuel J.
Computer Vision and Pattern Recognition
Materials Science
Feature foundation models - usually vision transformers - offer rich semantic descriptors of images, useful for downstream tasks such as (interactive) segmentation and object detection. For computational efficiency these descriptors are often patch-based, and so struggle to represent the fine features often present in micrographs; they also struggle with the large image sizes present in materials and biological image analysis. In this work, we train a convolutional neural network to upsample low-resolution (i.e, large patch size) foundation model features with reference to the input image. We apply this upsampler network (without any further training) to efficiently featurise and then segment a variety of microscopy images, including plant cells, a lithium-ion battery cathode and organic crystals. The richness of these upsampled features admits separation of hard to segment phases, like hairline cracks. We demonstrate that interactive segmentation with these deep features produces high-quality segmentations far faster and with far fewer labels than training or finetuning a more traditional convolutional network.
title Maybe you don't need a U-Net: convolutional feature upsampling for materials micrograph segmentation
topic Computer Vision and Pattern Recognition
Materials Science
url https://arxiv.org/abs/2508.21529