Self-supervised monocular image depth learning and confidence estimation
dc.contributor.author | Chen, Long | |
dc.contributor.author | Tang, Wen | |
dc.contributor.author | Wan, Tao Ruan | |
dc.contributor.author | John, Nigel W. | |
dc.date.accessioned | 2020-01-30T09:23:00Z | |
dc.date.available | 2020-01-30T09:23:00Z | |
dc.identifier.citation | Chen, L., Tang, W., Wan, T. R., & John, N. W., (2019). Self-supervised monocular image depth learning and confidence estimation. Neurocomputing, 381, 272-281. | en_US |
dc.identifier.uri | http://hdl.handle.net/10034/623135 | |
dc.description.abstract | We present a novel self-supervised framework for monocular image depth learning and confidence estimation. Our framework reduces the amount of ground truth annotation data required for training Convolutional Neural Networks (CNNs), which is often a challenging problem for the fast deployment of CNNs in many computer vision tasks. Our DepthNet adopts a novel fully differential patch-based cost function through the Zero-Mean Normalized Cross Correlation (ZNCC) to take multi-scale patches as matching and learning strategies. This approach greatly increases the accuracy and robustness of the depth learning. Whilst the proposed patch-based cost function naturally provides a 0-to-1 confidence, it is then used to self-supervise the training of a parallel network for confidence map learning and estimation by exploiting the fact that ZNCC is a normalized measure of similarity which can be approximated as the confidence of the depth estimation. Therefore, the proposed corresponding confidence map learning and estimation operate in a self-supervised manner and is a parallel network to the DepthNet. Evaluation on the KITTI depth prediction evaluation dataset and Make3D dataset show that our method outperforms the state-of-the-art results. | en_US |
dc.publisher | Elsevier | en_US |
dc.relation.url | https://www.sciencedirect.com/science/article/pii/S0925231219316388 | en_US |
dc.rights.uri | https://creativecommons.org/licenses/by-nc-nd/4.0/ | en_US |
dc.subject | Monocular depth estimation | en_US |
dc.subject | Deep convolutional neural networks | en_US |
dc.subject | Confidence map | en_US |
dc.title | Self-supervised monocular image depth learning and confidence estimation | en_US |
dc.type | Article | en_US |
dc.contributor.department | Bournemouth University; University of Bradford; University of Chester | en_US |
dc.identifier.journal | Neurocomputing | en_US |
or.grant.openaccess | Yes | en_US |
rioxxterms.funder | unfunded | en_US |
rioxxterms.identifier.project | unfunded | en_US |
rioxxterms.version | AM | en_US |
rioxxterms.versionofrecord | https://doi.org/10.1016/j.neucom.2019.11.038 | en_US |
rioxxterms.licenseref.startdate | 2020-12-04 | |
rioxxterms.publicationdate | 2019-12-04 | |
dc.dateAccepted | 2019-11-23 | |
dc.date.deposited | 2020-01-30 | en_US |
dc.indentifier.issn | 0925-2312 | en_US |