Loading [MathJax]/extensions/tex2jax.js

CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

Details

Original languageEnglish
Title of host publication2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2070-2079
Number of pages10
ISBN (electronic)978-1-7281-5023-9
ISBN (print)978-1-7281-5024-6
Publication statusPublished - Oct 2019
Event2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW) - Seoul, Korea, Republic of
Duration: 27 Oct 201928 Oct 2019

Publication series

NameIEEE International Conference on Computer Vision workshops
ISSN (Print)2473-9936
ISSN (electronic)2473-9944

Abstract

Due to its capability to identify erroneous disparity assignments in dense stereo matching, confidence estimation is beneficial for a wide range of applications, e.g. autonomous driving, which needs a high degree of confidence as mandatory prerequisite. Especially, the introduction of deep learning based methods resulted in an increasing popularity of this field in recent years, caused by a significantly improved accuracy. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, not taking into account the corresponding 3-dimensional cost volumes. However, it was already demonstrated that with conventional methods based on hand-crafted features this additional information can be used to further increase the accuracy. In order to combine the advantages of deep learning and cost volume based features, in this paper, we propose a novel Convolutional Neural Network (CNN) architecture to directly learn features for confidence estimation from volumetric 3D data. An extensive evaluation on three datasets using three common dense stereo matching techniques demonstrates the generality and state-of-the-art accuracy of the proposed method.

Keywords

    Confidence estimation, Convolutional neural network, Dense stereo matching

ASJC Scopus subject areas

Cite this

CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching. / Mehltretter, Max; Heipke, Christian.
2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Institute of Electrical and Electronics Engineers Inc., 2019. p. 2070-2079 9021993 (IEEE International Conference on Computer Vision workshops).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Mehltretter, M & Heipke, C 2019, CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching. in 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)., 9021993, IEEE International Conference on Computer Vision workshops, Institute of Electrical and Electronics Engineers Inc., pp. 2070-2079, 2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW), Seoul, Korea, Republic of, 27 Oct 2019. https://doi.org/10.48550/arXiv.1905.07287, https://doi.org/10.1109/ICCVW.2019.00262
Mehltretter, M., & Heipke, C. (2019). CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching. In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW) (pp. 2070-2079). Article 9021993 (IEEE International Conference on Computer Vision workshops). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.48550/arXiv.1905.07287, https://doi.org/10.1109/ICCVW.2019.00262
Mehltretter M, Heipke C. CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching. In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Institute of Electrical and Electronics Engineers Inc. 2019. p. 2070-2079. 9021993. (IEEE International Conference on Computer Vision workshops). doi: 10.48550/arXiv.1905.07287, 10.1109/ICCVW.2019.00262
Mehltretter, Max ; Heipke, Christian. / CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching. 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Institute of Electrical and Electronics Engineers Inc., 2019. pp. 2070-2079 (IEEE International Conference on Computer Vision workshops).
Download
@inproceedings{5ae652d4e3ed44d3931ad7305a25e9c5,
title = "CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching",
abstract = "Due to its capability to identify erroneous disparity assignments in dense stereo matching, confidence estimation is beneficial for a wide range of applications, e.g. autonomous driving, which needs a high degree of confidence as mandatory prerequisite. Especially, the introduction of deep learning based methods resulted in an increasing popularity of this field in recent years, caused by a significantly improved accuracy. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, not taking into account the corresponding 3-dimensional cost volumes. However, it was already demonstrated that with conventional methods based on hand-crafted features this additional information can be used to further increase the accuracy. In order to combine the advantages of deep learning and cost volume based features, in this paper, we propose a novel Convolutional Neural Network (CNN) architecture to directly learn features for confidence estimation from volumetric 3D data. An extensive evaluation on three datasets using three common dense stereo matching techniques demonstrates the generality and state-of-the-art accuracy of the proposed method.",
keywords = "Confidence estimation, Convolutional neural network, Dense stereo matching",
author = "Max Mehltretter and Christian Heipke",
note = "Funding information: This work was supported by the German Research Foundation (DFG) as a part of the Research Training Group i.c.sens [GRK2159], the MOBILISE initiative of the Leibniz University Hannover and TU Braunschweig and by the NVIDIA Corporation with the donation of the Titan V GPU used for this research.; 2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW), ICCVW ; Conference date: 27-10-2019 Through 28-10-2019",
year = "2019",
month = oct,
doi = "10.48550/arXiv.1905.07287",
language = "English",
isbn = "978-1-7281-5024-6",
series = "IEEE International Conference on Computer Vision workshops",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "2070--2079",
booktitle = "2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)",
address = "United States",

}

Download

TY - GEN

T1 - CNN-based Cost Volume Analysis as Confidence Measure for Dense Matching

AU - Mehltretter, Max

AU - Heipke, Christian

N1 - Funding information: This work was supported by the German Research Foundation (DFG) as a part of the Research Training Group i.c.sens [GRK2159], the MOBILISE initiative of the Leibniz University Hannover and TU Braunschweig and by the NVIDIA Corporation with the donation of the Titan V GPU used for this research.

PY - 2019/10

Y1 - 2019/10

N2 - Due to its capability to identify erroneous disparity assignments in dense stereo matching, confidence estimation is beneficial for a wide range of applications, e.g. autonomous driving, which needs a high degree of confidence as mandatory prerequisite. Especially, the introduction of deep learning based methods resulted in an increasing popularity of this field in recent years, caused by a significantly improved accuracy. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, not taking into account the corresponding 3-dimensional cost volumes. However, it was already demonstrated that with conventional methods based on hand-crafted features this additional information can be used to further increase the accuracy. In order to combine the advantages of deep learning and cost volume based features, in this paper, we propose a novel Convolutional Neural Network (CNN) architecture to directly learn features for confidence estimation from volumetric 3D data. An extensive evaluation on three datasets using three common dense stereo matching techniques demonstrates the generality and state-of-the-art accuracy of the proposed method.

AB - Due to its capability to identify erroneous disparity assignments in dense stereo matching, confidence estimation is beneficial for a wide range of applications, e.g. autonomous driving, which needs a high degree of confidence as mandatory prerequisite. Especially, the introduction of deep learning based methods resulted in an increasing popularity of this field in recent years, caused by a significantly improved accuracy. Despite this remarkable development, most of these methods rely on features learned from disparity maps only, not taking into account the corresponding 3-dimensional cost volumes. However, it was already demonstrated that with conventional methods based on hand-crafted features this additional information can be used to further increase the accuracy. In order to combine the advantages of deep learning and cost volume based features, in this paper, we propose a novel Convolutional Neural Network (CNN) architecture to directly learn features for confidence estimation from volumetric 3D data. An extensive evaluation on three datasets using three common dense stereo matching techniques demonstrates the generality and state-of-the-art accuracy of the proposed method.

KW - Confidence estimation

KW - Convolutional neural network

KW - Dense stereo matching

UR - http://www.scopus.com/inward/record.url?scp=85078929241&partnerID=8YFLogxK

U2 - 10.48550/arXiv.1905.07287

DO - 10.48550/arXiv.1905.07287

M3 - Conference contribution

AN - SCOPUS:85078929241

SN - 978-1-7281-5024-6

T3 - IEEE International Conference on Computer Vision workshops

SP - 2070

EP - 2079

BT - 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW)

Y2 - 27 October 2019 through 28 October 2019

ER -

By the same author(s)