Dealing with Topological Information within a Fully Convolutional Neural Network
Computer Vision and Pattern Recognition
2019-06-28 v1 Image and Video Processing
Abstract
A fully convolutional neural network has a receptive field of limited size and therefore cannot exploit global information, such as topological information. A solution is proposed in this paper to solve this problem, based on pre-processing with a geodesic operator. It is applied to the segmentation of histological images of pigmented reconstructed epidermis acquired via Whole Slide Imaging.
Cite
@article{arxiv.1906.11600,
title = {Dealing with Topological Information within a Fully Convolutional Neural Network},
author = {Etienne Decencière and Santiago Velasco-Forero and Fu Min and Juanjuan Chen and Hélène Burdin and Gervais Gauthier and Bruno Laÿ and Thomas Bornschloegl and Thérèse Baldeweck},
journal= {arXiv preprint arXiv:1906.11600},
year = {2019}
}
Comments
International Conference on Advanced Concepts for Intelligent Vision Systems (ACIVS 2018)