Semantic image segmentation with fused CNN features

被引：10

作者：

Geng H.-Q. ^{[1
]}

Zhang H. ^{[1
]}

Xue Y.-B. ^{[1
]}

Zhou M. ^{[1
]}

Xu G.-P. ^{[1
]}

Gao Z. ^{[1
]}

机构：

[1] Key Laboratory of Computer Vision and System of Ministry of Education, Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology, Tianjin University of Technology, Tianjin

来源：

Optoelectronics Letters | 2017年 / 13卷 / 05期

基金：

中国国家自然科学基金;

关键词：

Neural networks - Random processes - Image representation - Semantics - Semantic Segmentation;

D O I：

10.1007/s11801-017-7086-6

中图分类号：

学科分类号：

摘要：

Semantic image segmentation is a task to predict a category label for every image pixel. The key challenge of it is to design a strong feature representation. In this paper, we fuse the hierarchical convolutional neural network (CNN) features and the region-based features as the feature representation. The hierarchical features contain more global information, while the region-based features contain more local information. The combination of these two kinds of features significantly enhances the feature representation. Then the fused features are used to train a softmax classifier to produce per-pixel label assignment probability. And a fully connected conditional random field (CRF) is used as a post-processing method to improve the labeling consistency. We conduct experiments on SIFT flow dataset. The pixel accuracy and class accuracy are 84.4% and 34.86%, respectively. © 2017, Tianjin University of Technology and Springer-Verlag GmbH Germany.

引用

页码：381 / 385

页数：4