Real-time monocular object SLAM

被引:95
作者
Galvez-Lopez, Dorian [1 ]
Salas, Marta [1 ]
Tardos, Juan D. [1 ]
Montiel, J. M. M. [1 ]
机构
[1] Univ Zaragoza, Inst Invest Ingn Aragon I3A, Zaragoza 50018, Spain
关键词
Object slam; Object recognition; RECOGNITION; LOCALIZATION; SEARCH;
D O I
10.1016/j.robot.2015.08.009
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We present a real-time object-based SLAM system that leverages the largest object database to date. Our approach comprises two main components: (1) a monocular SLAM algorithm that exploits object rigidity constraints to improve the map and find its real scale, and (2) a novel object recognition algorithm based on bags of binary words, which provides live detections with a database of 500 3D objects. The two components work together and benefit each other: the SLAM algorithm accumulates information from the observations of the objects, anchors object features to especial map landmarks and sets constrains on the optimization. At the same time, objects partially or fully located within the map are used as a prior to guide the recognition algorithm, achieving higher recall. We evaluate our proposal on five real environments showing improvements on the accuracy of the map and efficiency with respect to other state-of-the-art techniques. (C) 2015 Elsevier B.V. All rights reserved.
引用
收藏
页码:435 / 449
页数:15
相关论文
共 49 条
[1]  
[Anonymous], 2012, P INT C INT ROB SYST
[2]  
[Anonymous], 2006, 2006 IEEE COMP SOC C
[3]  
[Anonymous], 2013, IEEE C COMP VIS PATT
[4]  
[Anonymous], IEEE COMPUTER VISION
[5]  
[Anonymous], 2001, Robotica, DOI DOI 10.1017/S0263574700223217
[6]  
[Anonymous], 2011, IEEERSJ INT C INTELL
[7]  
[Anonymous], P EUR C COMP VIS ECC
[8]  
[Anonymous], IEEE T PATTERN ANAL
[9]  
Bao S. Y., 2011, 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), P2025, DOI 10.1109/CVPR.2011.5995462
[10]  
Bao SY, 2012, PROC CVPR IEEE, P2703, DOI 10.1109/CVPR.2012.6247992