Paper Info
Title
On-Board monocular vision system pose estimation through a dense optical flow
Abstract
This paper presents a robust technique for estimating on-board monocular vision system pose. The proposed approach is based on a dense optical flow that is robust against shadows, reflections and illumination changes. A RANSAC based scheme is used to cope with the outliers in the optical flow. The proposed technique is intended to be used in driver assistance systems for applications such as obstacle or pedestrian detection. Experimental results on different scenarios, both from synthetic and real sequences, shows usefulness of the proposed approach.
Year
DOI
Venue
2010
10.1007/978-3-642-13772-3_24
ICIAR
Keywords
Field
DocType
illumination change,proposed technique,driver assistance system,different scenario,optical flow,on-board monocular vision system,dense optical flow,robust technique,pose estimation,monocular vision
Monocular vision,Obstacle,Computer vision,RANSAC,Computer science,Advanced driver assistance systems,Pose,Artificial intelligence,Intelligent transportation system,Optical flow,Pedestrian detection
Conference
Volume
ISSN
ISBN
6111
0302-9743
3-642-13771-7
Citations 
PageRank 
References 
3
0.42
9
Authors
2
Name
Order
Citations
PageRank
Naveen Onkarappa1214.11
Angel Domingo Sappa256533.54