Indoor positioning using convolution neural network to regress camera pose

Jing Mei Ciou, E. H.C. Lu

Research output: Contribution to journalConference article

Abstract

In recent years, the issue of indoor positioning has become more and more popular and attracted more attention. Under the absence of GNSS, how to more accurately position is one of the challenges on the positioning technology. Camera positioning can be calculated by image and objects. Therefore, this study focuses on locating the user's camera position, but how to calculate the camera position efficiently is a very challenging problem. With the rapid development of neural network in image recognition, computer can not only process images quickly, but also achieve good results. Convolution Neural Network (CNN) can sense the local area of the image and find some high-resolution local features. These basic features are likely to form the basis of human vision and become an effective means to improve the recognition rate. We use a 23-layer convolutional neural network architecture and set different sizes of input images to train the end-to-end task of location recognition to regress the camera's position and direction. We choose the sites where are the underground parking lot for the experiment. Compared with other indoor environments such as chess, office and kitchen, the condition of this place is very severe. Therefore, how to design algorithms to train and exclude dynamic objects using neural networks is very exploratory. The experimental results show that our proposed solution can effectively reduce the error of indoor positioning.

Original languageEnglish
Pages (from-to)1289-1294
Number of pages6
JournalInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
Volume42
Issue number2/W13
DOIs
Publication statusPublished - 2019 Jun 4
Event4th ISPRS Geospatial Week 2019 - Enschede, Netherlands
Duration: 2019 Jun 102019 Jun 14

Fingerprint

Convolution
neural network
positioning
Cameras
Neural networks
train
Kitchens
Image recognition
Parking
Network architecture
GNSS
parking
experiment
Experiments

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Geography, Planning and Development

Cite this

@article{5c45d7e25f1847059e103ed8f2f68c65,
title = "Indoor positioning using convolution neural network to regress camera pose",
abstract = "In recent years, the issue of indoor positioning has become more and more popular and attracted more attention. Under the absence of GNSS, how to more accurately position is one of the challenges on the positioning technology. Camera positioning can be calculated by image and objects. Therefore, this study focuses on locating the user's camera position, but how to calculate the camera position efficiently is a very challenging problem. With the rapid development of neural network in image recognition, computer can not only process images quickly, but also achieve good results. Convolution Neural Network (CNN) can sense the local area of the image and find some high-resolution local features. These basic features are likely to form the basis of human vision and become an effective means to improve the recognition rate. We use a 23-layer convolutional neural network architecture and set different sizes of input images to train the end-to-end task of location recognition to regress the camera's position and direction. We choose the sites where are the underground parking lot for the experiment. Compared with other indoor environments such as chess, office and kitchen, the condition of this place is very severe. Therefore, how to design algorithms to train and exclude dynamic objects using neural networks is very exploratory. The experimental results show that our proposed solution can effectively reduce the error of indoor positioning.",
author = "Ciou, {Jing Mei} and Lu, {E. H.C.}",
year = "2019",
month = "6",
day = "4",
doi = "10.5194/isprs-archives-XLII-2-W13-1289-2019",
language = "English",
volume = "42",
pages = "1289--1294",
journal = "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives",
issn = "1682-1750",
number = "2/W13",

}

Indoor positioning using convolution neural network to regress camera pose. / Ciou, Jing Mei; Lu, E. H.C.

In: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, Vol. 42, No. 2/W13, 04.06.2019, p. 1289-1294.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Indoor positioning using convolution neural network to regress camera pose

AU - Ciou, Jing Mei

AU - Lu, E. H.C.

PY - 2019/6/4

Y1 - 2019/6/4

N2 - In recent years, the issue of indoor positioning has become more and more popular and attracted more attention. Under the absence of GNSS, how to more accurately position is one of the challenges on the positioning technology. Camera positioning can be calculated by image and objects. Therefore, this study focuses on locating the user's camera position, but how to calculate the camera position efficiently is a very challenging problem. With the rapid development of neural network in image recognition, computer can not only process images quickly, but also achieve good results. Convolution Neural Network (CNN) can sense the local area of the image and find some high-resolution local features. These basic features are likely to form the basis of human vision and become an effective means to improve the recognition rate. We use a 23-layer convolutional neural network architecture and set different sizes of input images to train the end-to-end task of location recognition to regress the camera's position and direction. We choose the sites where are the underground parking lot for the experiment. Compared with other indoor environments such as chess, office and kitchen, the condition of this place is very severe. Therefore, how to design algorithms to train and exclude dynamic objects using neural networks is very exploratory. The experimental results show that our proposed solution can effectively reduce the error of indoor positioning.

AB - In recent years, the issue of indoor positioning has become more and more popular and attracted more attention. Under the absence of GNSS, how to more accurately position is one of the challenges on the positioning technology. Camera positioning can be calculated by image and objects. Therefore, this study focuses on locating the user's camera position, but how to calculate the camera position efficiently is a very challenging problem. With the rapid development of neural network in image recognition, computer can not only process images quickly, but also achieve good results. Convolution Neural Network (CNN) can sense the local area of the image and find some high-resolution local features. These basic features are likely to form the basis of human vision and become an effective means to improve the recognition rate. We use a 23-layer convolutional neural network architecture and set different sizes of input images to train the end-to-end task of location recognition to regress the camera's position and direction. We choose the sites where are the underground parking lot for the experiment. Compared with other indoor environments such as chess, office and kitchen, the condition of this place is very severe. Therefore, how to design algorithms to train and exclude dynamic objects using neural networks is very exploratory. The experimental results show that our proposed solution can effectively reduce the error of indoor positioning.

UR - http://www.scopus.com/inward/record.url?scp=85067480004&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067480004&partnerID=8YFLogxK

U2 - 10.5194/isprs-archives-XLII-2-W13-1289-2019

DO - 10.5194/isprs-archives-XLII-2-W13-1289-2019

M3 - Conference article

AN - SCOPUS:85067480004

VL - 42

SP - 1289

EP - 1294

JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives

SN - 1682-1750

IS - 2/W13

ER -