DocumentCode :
3748741
Title :
Lost Shopping! Monocular Localization in Large Indoor Spaces
Author :
Shenlong Wang;Sanja Fidler;Raquel Urtasun
fYear :
2015
Firstpage :
2695
Lastpage :
2703
Abstract :
In this paper we propose a novel approach to localization in very large indoor spaces (i.e., 200+ store shopping malls) that takes a single image and a floor plan of the environment as input. We formulate the localization problem as inference in a Markov random field, which jointly reasons about text detection (localizing shop´s names in the image with precise bounding boxes), shop facade segmentation, as well as camera´s rotation and translation within the entire shopping mall. The power of our approach is that it does not use any prior information about appearance and instead exploits text detections corresponding to the shop names. This makes our method applicable to a variety of domains and robust to store appearance variation across countries, seasons, and illumination conditions. We demonstrate the performance of our approach in a new dataset we collected of two very large shopping malls, and show the power of holistic reasoning.
Keywords :
"Cameras","Three-dimensional displays","Layout","Image segmentation","Robustness","Lighting","Global Positioning System"
Publisher :
ieee
Conference_Titel :
Computer Vision (ICCV), 2015 IEEE International Conference on
Electronic_ISBN :
2380-7504
Type :
conf
DOI :
10.1109/ICCV.2015.309
Filename :
7410666
Link To Document :
بازگشت