Automatic Discovery and Geotagging of Objects from Street View Imagery
File Type:
PDFItem Type:
Journal ArticleDate:
2018Access:
openAccessCitation:
Kenny, E., Krylov, V., Dahyot, R. Automatic Discovery and Geotagging of Objects from Street View Imagery, Remote Sensing, 2018, 10, 5, 661Download Item:
remotesensing-10-00661 (1).pdf (PDF) 3.511Mb
Abstract:
Many applications, such as autonomous navigation, urban planning, and asset monitoring, rely on the availability of accurate information about objects and their geolocations. In this paper, we propose the automatic detection and computation of the coordinates of recurring stationary objects of interest using street view imagery. Our processing pipeline relies on two fully convolutional neural networks: the first segments objects in the images, while the second estimates their distance from the camera. To geolocate all the detected objects coherently we propose a novel custom Markov random field model to estimate the objects’ geolocation. The novelty of the resulting pipeline is the combined use of monocular depth estimation and triangulation to enable automatic mapping of complex scenes with the simultaneous presence of multiple, visually similar objects of interest. We validate experimentally the effectiveness of our approach on two object classes: traffic lights and telegraph poles. The experiments report high object recall rates and position precision of approximately 2 m, which is approaching the precision of single-frequency GPS receivers.
Sponsor
Grant Number
Science Foundation Ireland (SFI)
13/RC/2106
Science Foundation Ireland
13/RC/2106
Author's Homepage:
http://people.tcd.ie/dahyotrhttp://people.tcd.ie/krylovv
http://people.tcd.ie/ekenny
Description:
PUBLISHEDType of material:
Journal ArticleCollections:
Series/Report no:
Remote Sensing;10;
5;
Availability:
Full text availableKeywords:
Object geolocation, Object mapping, Street view imagery, Markov random fields, Traffic lights, Telecom assets, GPS estimationSubject (TCD):
Smart & Sustainable Planet , Telecommunications , COLOUR OBJECT DETECTION , Computer Vision and Image Processing , Data Analysis , Information technology in education , deep learningDOI:
http://dx.doi.org/10.3390/rs10050661Licences: