Dealing with Topological Information within a Fully Convolutional Neural Network

06/27/2019

∙

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.

READ FULL TEXT

Dealing with Topological Information within a Fully Convolutional Neural Network

Saliency Detection With Fully Convolutional Neural Network

Translated Skip Connections – Expanding the Receptive Fields of Fully Convolutional Neural Networks

Prostate Segmentation from Ultrasound Images using Residual Fully Convolutional Network

Dilated Fully Convolutional Neural Network for Depth Estimation from a Single Image

Detection of Single Grapevine Berries in Images Using Fully Convolutional Neural Networks

Switchblade – a Neural Network for Hard 2D Tasks

Learning Spatial-Aware Regressions for Visual Tracking

Dealing with Topological Information within a Fully Convolutional Neural Network

Related Research

Saliency Detection With Fully Convolutional Neural Network

Translated Skip Connections – Expanding the Receptive Fields of Fully Convolutional Neural Networks

Prostate Segmentation from Ultrasound Images using Residual Fully Convolutional Network

Dilated Fully Convolutional Neural Network for Depth Estimation from a Single Image

Detection of Single Grapevine Berries in Images Using Fully Convolutional Neural Networks

Switchblade – a Neural Network for Hard 2D Tasks

Learning Spatial-Aware Regressions for Visual Tracking