Inpainting Networks Learn to Separate Cells in Microscopy Images

janelia7_blocks-janelia7_biblio_header | block

British Machine Vision Conference. 2020 Sep

Inpainting Networks Learn to Separate Cells in Microscopy Images Funke Lab

Wolf S, Hamprecht FA, Funke J

janelia7_blocks-janelia7_biblio_abstract | block

Abstract

Deep neural networks trained to inpaint partially occluded images show a deep understanding of image composition and have even been shown to remove objects from images convincingly. In this work, we investigate how this implicit knowledge of image composition can be be used to separate cells in densely populated microscopy images. We propose a measure for the independence of two image regions given a fully self-supervised inpainting network and separate objects by maximizing this independence. We evaluate our method on two cell segmentation datasets and show that cells can be separated completely unsupervised. Furthermore, combined with simple foreground detection, our method yields instance segmentation of similar quality to fully supervised methods.

node:body | entity_field

Main Menu (Mobile)- Block

Main Menu - Block

Inpainting Networks Learn to Separate Cells in Microscopy Images

Abstract

Janelia Authors