KE, Rihuan; BUGEAU, Aurélie; PAPADAKIS, Nicolas; SCHUETZ, Peter; SCHÖNLIEB, Carola-Bibiane

Métadonnées

Afficher la notice complète

Licence d’utilisation du document

KE, Rihuan
Department of Applied Mathematics and Theoretical Physics [DAMTP]

BUGEAU, Aurélie
Laboratoire Bordelais de Recherche en Informatique [LaBRI]

PAPADAKIS, Nicolas
Institut de Mathématiques de Bordeaux [IMB]

Langue

Communication dans un congrès

Ce document a été publié dans

ECCV Workshop on BioImage Computing (BIC'20), 2020-08-23, Glasgow. 2020-08-23p. 411-428

Résumé en anglais

The need for labour intensive pixel-wise annotation is a major limitation of many fully supervised learning methods for segmenting bioimages that can contain numerous object instances with thin separations. In this paper, we introduce a deep convolutional neural network for microscopy image segmentation. Annotation issues are circumvented by letting the network being trainable on coarse labels combined with only a very small number of images with pixel-wise annotations. We call this new labelling strategy `lazy' labels. Image segmentation is stratified into three connected tasks: rough inner region detection, object separation and pixel-wise segmentation. These tasks are learned in an end-to-end multi-task learning framework. The method is demonstrated on two microscopy datasets, where we show that the model gives accurate segmentation results even if exact boundary labels are missing for a majority of annotated data. It brings more flexibility and efficiency for training deep neural networks that are data hungry and is applicable to biomedical images with poor contrast at the object boundaries or with diverse textures and repeated patterns.< Réduire

Mots clés en anglais

Multi-task learning

Convolutional neural networks

Image segmentation

URI

https://oskar-bordeaux.fr/handle/20.500.12278/191857

Projet Européen

Nonlocal Methods for Arbitrary Data Sources

Origine

Importé de hal

Unités de recherche

Institut de Mathématiques de Bordeaux (IMB) - UMR 5251