Semantic Segmentation via Global Convolutional Network and Concatenated Feature Maps

Chuan Kai Wang, Long Wen Chang

2019

Abstract

Most of the segmentation CNNs (convolutional neural network) based on the ResNet. Recently, Huang et al. introduced a new classification CNN called DenseNet. Then Jégou et al. used a sequence of building blocks for DenseNet to build their semantic segmentation CNN, called FC-DenseNet, and achieved state-of-the-art results on CamVid dataset. In this paper, we implement the design concept of DenseNet into a ResNet-based semantic segmentation CNN called Global Convolutional Network (GCN) and build our own network by switching every identity mapping operation of the decoder network in GCN to a concatenation operation. Our network uses less computational resources than FC-DenseNet to obtain a mean IoU score of 69.34% on CamVid dataset, and surpass the 66.9% obtained in the paper of FC-DenseNet.

Download


Paper Citation


in Harvard Style

Wang C. and Chang L. (2019). Semantic Segmentation via Global Convolutional Network and Concatenated Feature Maps.In Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-351-3, pages 292-297. DOI: 10.5220/0007251002920297


in Bibtex Style

@conference{icpram19,
author={Chuan Wang and Long Chang},
title={Semantic Segmentation via Global Convolutional Network and Concatenated Feature Maps},
booktitle={Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2019},
pages={292-297},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007251002920297},
isbn={978-989-758-351-3},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Semantic Segmentation via Global Convolutional Network and Concatenated Feature Maps
SN - 978-989-758-351-3
AU - Wang C.
AU - Chang L.
PY - 2019
SP - 292
EP - 297
DO - 10.5220/0007251002920297