Semantic image segmentation with fused CNN features |
| |
Authors: | GENG Hui-qiang ZHANG Hu XUE Yan-bing ZHOU Mian XU Guang-ping GAO Zan |
| |
Affiliation: | 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 300384, China,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 300384, China,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 300384, China,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 300384, China,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 300384, China and 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 300384, China |
| |
Abstract: | 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. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
| 点击此处可从《光电子快报》浏览原始摘要信息 |
|
点击此处可从《光电子快报》下载全文 |
|