• 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

作者:

Zuo, Guoyu (Zuo, Guoyu.) (学者:左国玉) | Zheng, Tao (Zheng, Tao.) | Xu, Zichen (Xu, Zichen.) | Gong, Daoxiong (Gong, Daoxiong.)

收录:

EI

摘要:

This paper proposes a fine semantic mapping method using dense segmentation network (DS-Net) to obtain good performance of semantic mapping fusion, in which the semantic segmentation network (DS-Net) is constructed based on the idea of DenseNet's dense connection. First, the RGB image and the depth image are used to generate a dense indoor scene map via the state-of-the-art dense SLAM (ElasticFusion). Then, semantic segmentation are precisely performed on the input RGB image via DS-Net. Finally, the long-term correspondence between the landmarks and the indoor scene map is established using the continuous frames both in the visual odometer and loop detection, and the final fused semantic map is obtained by integrating semantic predictions of the RGB-D video frames of multiple angles with the indoor scene map. Experiments were performed on the NYUv2 and CIFAR10 datasets and our laboratory environments. Results show shows that our method can reduce the error of dense map construction and obtain good semantic segmentation performance. © 2019 IEEE.

关键词:

Agricultural robots Biomimetics Image segmentation Mapping Robotics Semantics Semantic Web

作者机构:

  • [ 1 ] [Zuo, Guoyu]Beijing University of Technology, Faculty of Information Technology, Beijing; 100124, China
  • [ 2 ] [Zheng, Tao]Beijing University of Technology, Faculty of Information Technology, Beijing; 100124, China
  • [ 3 ] [Xu, Zichen]Beijing University of Technology, Faculty of Information Technology, Beijing; 100124, China
  • [ 4 ] [Gong, Daoxiong]Beijing University of Technology, Faculty of Information Technology, Beijing; 100124, China

通讯作者信息:

  • 左国玉

    [zuo, guoyu]beijing university of technology, faculty of information technology, beijing; 100124, china

电子邮件地址:

查看成果更多字段

相关关键词:

相关文章:

来源 :

年份: 2019

页码: 1969-1974

语种: 英文

被引次数:

WoS核心集被引频次: 0

SCOPUS被引频次: 1

ESI高被引论文在榜: 0 展开所有

万方被引频次:

中文被引频次:

近30日浏览量: 2

归属院系:

在线人数/总访问数:104/3601080
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司