Multiscale Feature Extractors for Stereo Matching Cost Computation

We propose four efficient feature extractors based on convolutional neural networks for stereo matching cost computation. Two of them generate multiscale features with diverse receptive field sizes. These multiscale features are used to compute the corresponding multiscale matching costs. We then de...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2018-01, Vol.6, p.27971-27983
Main Authors: Kim, Kyung-Rae, Koh, Yeong Jun, Kim, Chang-Su
Format: Article
Language:English
Subjects:
Algorithms
Artificial neural networks
Computational complexity
Computational modeling
Convolutional neural networks
Costs
Data mining
Electrical engineering
Feature extraction
Matching
matching cost computation
multiscale feature extraction
Reliability
Stereo matching
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose four efficient feature extractors based on convolutional neural networks for stereo matching cost computation. Two of them generate multiscale features with diverse receptive field sizes. These multiscale features are used to compute the corresponding multiscale matching costs. We then determine an optimal cost by combining the multiscale costs using edge information. On the other hand, the other two feature extractors produce uni-scale features by combining multiscale features directly through fully connected layers. Finally, after obtaining matching costs using one of the four extractors, we determine optimal disparities based on the cross-based cost aggregation and the semiglobal matching. Extensive experiments on the Middlebury stereo data sets demonstrate the effectiveness and efficiency of the proposed algorithm. Specifically, the proposed algorithm provides competitive matching performance with the state of the arts, while demanding lower computational complexity.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2018.2838442