Robust Real-time RGB-D Visual Odometry in Dynamic Environments via Rigid Motion Model

被引:0
|
作者
Lee, Sangil [1 ]
Son, Clark Youngdong [1 ]
Kim, H. Jin [1 ]
机构
[1] Seoul Natl Univ, Dept Mech & Aerosp Engn, Automat & Syst Res Inst, Seoul 08826, South Korea
来源
2019 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) | 2019年
关键词
D O I
10.1109/iros40897.2019.8968208
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In the paper, we propose a robust real-time visual odometry in dynamic environments via rigid-motion model updated by scene flow. The proposed algorithm consists of spatial motion segmentation and temporal motion tracking. The spatial segmentation first generates several motion hypotheses by using a grid-based scene flow and clusters the extracted motion hypotheses, separating objects that move independently of one another. Further, we use a dual-mode motion model to consistently distinguish between the static and dynamic parts in the temporal motion tracking stage. Finally, the proposed algorithm estimates the pose of a camera by taking advantage of the region classified as static parts. In order to evaluate the performance of visual odometry under the existence of dynamic rigid objects, we use self-collected dataset containing RGB-D images and motion capture data for ground-truth. We compare our algorithm with state-of-the-art visual odometry algorithms. The validation results suggest that the proposed algorithm can estimate the pose of a camera robustly and accurately in dynamic environments.
引用
收藏
页码:6891 / 6898
页数:8
相关论文
共 50 条
  • [1] Robust Real-Time Visual Odometry for Dense RGB-D Mapping
    Whelan, Thomas
    Johannsson, Hordur
    Kaess, Michael
    Leonard, John J.
    McDonald, John
    2013 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2013, : 5724 - 5731
  • [2] A robust visual odometry based on RGB-D camera in dynamic indoor environments
    Zhang, Fangfang
    Li, Qiyan
    Wang, Tingting
    Ma, Tianlei
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2021, 32 (04)
  • [3] Real-Time Visual Odometry from Dense RGB-D Images
    Steinbruecker, Frank
    Sturm, Juergen
    Cremers, Daniel
    2011 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCV WORKSHOPS), 2011,
  • [4] Dynamic RGB-D Visual Odometry
    Yang, Dongsheng
    Bi, Shusheng
    Cai, Yueri
    Zheng, Jingxiang
    Yuan, Chang
    2017 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (IEEE ROBIO 2017), 2017, : 941 - 946
  • [5] Efficient Compositional Approaches for Real-Time Robust Direct Visual Odometry from RGB-D Data
    Klose, Sebastian
    Heise, Philipp
    Knoll, Alois
    2013 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2013, : 1100 - 1106
  • [6] An evaluation of real-time RGB-D visual odometry algorithms on mobile devices
    Vincent Angladon
    Simone Gasparini
    Vincent Charvillat
    Tomislav Pribanić
    Tomislav Petković
    Matea Ðonlić
    Benjamin Ahsan
    Frédéric Bruel
    Journal of Real-Time Image Processing, 2019, 16 : 1643 - 1660
  • [7] An evaluation of real-time RGB-D visual odometry algorithms on mobile devices
    Angladon, Vincent
    Gasparini, Simone
    Charvillat, Vincent
    Pribanic, Tomislav
    Petkovic, Tomislav
    Donlic, Matea
    Ahsan, Benjamin
    Bruel, Frederic
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2019, 16 (05) : 1643 - 1660
  • [8] RGB-D Visual Odometry in Dynamic Environments Using Line Features
    Zhang H.
    Fang Z.
    Yang G.
    Jiqiren/Robot, 2019, 41 (01): : 75 - 82
  • [9] Real-time Visual Odometry for Autonomous MAV Navigation using RGB-D Camera
    Wang, Jiefei
    Garratt, Matthew
    Anavatti, Sreenatha
    Lin, Shanggang
    2015 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO), 2015, : 1353 - 1358
  • [10] Towards Real-time Semantic RGB-D SLAM in Dynamic Environments
    Ji, Tete
    Wang, Chen
    Xie, Lihua
    2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021), 2021, : 11175 - 11181