BibTeX
@MISC{Morales_real-worldstereo-analysis,
author = {Ino Morales and Simon Hermann and Reinhard Klette},
title = {Real-World Stereo-Analysis Evaluation},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
www.mi.auckland.ac.nz Abstract. Evaluation of stereo-analysis algorithms is usually done by analysing the performance of stereo matchers on data sets with available ground truth. The trade-off between precise results, obtained with this sort of evaluation, and the limited amount (in both, quantity and diversity) of data sets, needs to be considered if the algorithms are intended to be used for the analysis of real-world environments. The quality of the computed disparity maps is affected by common and unavoidable noise present in real-world images. In this chapter we discus an approach to objectively evaluate the performance of stereo-analysis algorithms using real-world image sequences. The lack of ground truth is tackled by incorporating an extra camera into a multi-view stereo camera system. The proposed, relatively simple hardware set-up can easily be reproduced such that test-data sets can be generated for specific applications. 1
Keyphrases
real-world stereo-analysis evaluation data set stereo-analysis algorithm limited amount ground truth stereo matcher real-world image sequence extra camera test-data set real-world environment precise result multi-view stereo camera system real-world image available ground truth computed disparity map specific application nz abstract unavoidable noise present
