Real-time 3D semantic map building in indoor scene

被引：0
|
作者：
Shan J. [1 ,2 ]
Li X. [1 ,2 ]
Zhang X. [1 ,2 ]
Jia S. [1 ,2 ]
机构：
[1] Faculty of Information Technology, Beijing University of Technology, Beijing
[2] Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing
来源：
Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument | 2019年 / 40卷 / 05期
关键词：
3D semantic map; Bayesian update; Closed-loop optimization; Image semantic segmentation; SLAM;
D O I：
10.19650/j.cnki.cjsi.J1904749
中图分类号：
学科分类号：
摘要：
Autonomous mapping for mobile robot is the premise of completing intelligent behavior. To improve the intelligence and intuitive user interaction of robot, maps are needed to achieve the semantics beyond geometry and appearance. This paper studies the 3D semantic map construction method, which fuses the pixel-level image semantic segmentation based on Deep Residual Networks(DRN) and Simultaneous Localization And Mapping(SLAM). Firstly, the combined median filter algorithmis used to restore the depth of the map. The improved Iterator Closest Point (ICP) algorithm is employed to estimate camera pose and loopback detection based on random ferns is proposed for 3D scene reconstruction. Then, the optimized DRN is utilized to achieve more accurate semantic prediction and segmentation. Finally, the predicted semantic classification labels are migrated to the 3D model by Bayesian based incremental transfer strategy to generate a globally consistent 3D semantic map. Experimental results show that the proposed method can build the real-time 3D semantic map in the real and complicated environment. © 2019, Science Press. All right reserved.
引用
收藏
页码：240 / 248
页数：8
相关论文
共 23 条
  • [1] Tang H.W., Sun W., Yang K., Et al., Grid map merging approach of multi-robot based on SURF feature, Journal of Electronic Measurement and Instrumentation, 31, 6, pp. 859-868, (2017)
  • [2] Schwertfeger S., Birk A., Map evaluation using matched topology graphs, Autonomous Robots, 40, 5, pp. 761-787, (2016)
  • [3] Nieto-Granda C., Rogers J.G., Trevor A.J.B., Et al., Semantic map partitioning in indoor environments usingregional analysis, International Conference on Intelligent Robots and Systems, pp. 1451-1456, (2010)
  • [4] Kostavelis I., Gasteratosa, Learning spatially semantic representations for cognitive robot navigation, Robotics and Autonomous Systems, 61, 12, pp. 1460-1475, (2013)
  • [5] Rituerto A., Murillo A.C., Guerrero J.J., Semantic labeling for indoor topological mapping using a wearablecatadioptric system, Robotics and Autonomous Systems, 62, 5, pp. 685-695, (2014)
  • [6] Wu H., Tian G., Xue Y., Et al., QR code based semantic map building in domestic semi-unknown environment, Pattern Recognition and Artificial Intelligence, 23, 4, pp. 464-470, (2010)
  • [7] Wu H., Tian G.H., Wang J.C., Et al., Three-dimensional grid semantic map building in unstructured indoor environment, Pattern Recognition and Artificial Intelligence, 4, pp. 564-572, (2012)
  • [8] Sheng W., Du J., Cheng Q., Et al., Robot semantic mapping through human activity recognition: A wearable sensing and computing approach, Robotics and Autonomous Systems, 68, pp. 47-58, (2015)
  • [9] Song S.R., Yu F., Zeng A., Semantic scene completion from a single depth image, International Conference on Computer Vision and Pattern Recognition, pp. 190-198, (2017)
  • [10] Li X.Z., Li S.Y., A system of real time mobile robot semantic map building, Chinese Journal of Scientific Instrument, 38, 11, pp. 2769-2778, (2017)
123