Tampere University of Technology

TUTCRIS Research Portal

DGC-Net: Dense geometric correspondence network

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

DGC-Net : Dense geometric correspondence network. / Melekhov, Iaroslav; Tiulpin, Aleksei; Sattler, Torsten; Pollefeys, Marc; Rahtu, Esa; Kannala, Juho.

2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019. IEEE, 2019. p. 1034-1042 (IEEE Winter Conference on Applications of Computer Vision).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Melekhov, I, Tiulpin, A, Sattler, T, Pollefeys, M, Rahtu, E & Kannala, J 2019, DGC-Net: Dense geometric correspondence network. in 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019. IEEE Winter Conference on Applications of Computer Vision, IEEE, pp. 1034-1042, IEEE Winter Conference on Applications of Computer Vision, Waikoloa Village, United States, 7/01/19. https://doi.org/10.1109/WACV.2019.00115

APA

Melekhov, I., Tiulpin, A., Sattler, T., Pollefeys, M., Rahtu, E., & Kannala, J. (2019). DGC-Net: Dense geometric correspondence network. In 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019 (pp. 1034-1042). (IEEE Winter Conference on Applications of Computer Vision). IEEE. https://doi.org/10.1109/WACV.2019.00115

Vancouver

Melekhov I, Tiulpin A, Sattler T, Pollefeys M, Rahtu E, Kannala J. DGC-Net: Dense geometric correspondence network. In 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019. IEEE. 2019. p. 1034-1042. (IEEE Winter Conference on Applications of Computer Vision). https://doi.org/10.1109/WACV.2019.00115

Author

Melekhov, Iaroslav ; Tiulpin, Aleksei ; Sattler, Torsten ; Pollefeys, Marc ; Rahtu, Esa ; Kannala, Juho. / DGC-Net : Dense geometric correspondence network. 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019. IEEE, 2019. pp. 1034-1042 (IEEE Winter Conference on Applications of Computer Vision).

Bibtex - Download

@inproceedings{8b7770bea65f4224966bae92726f4bb3,
title = "DGC-Net: Dense geometric correspondence network",
abstract = "This paper addresses the challenge of dense pixel correspondence estimation between two images. This problem is closely related to optical flow estimation task where Con-vNets (CNNs) have recently achieved significant progress. While optical flow methods produce very accurate results for the small pixel translation and limited appearance variation scenarios, they hardly deal with the strong geometric transformations that we consider in this work. In this paper, we propose a coarse-to-fine CNN-based framework that can leverage the advantages of optical flow approaches and extend them to the case of large transformations providing dense and subpixel accurate estimates. It is trained on synthetic transformations and demonstrates very good performance to unseen, realistic, data. Further, we apply our method to the problem of relative camera pose estimation and demonstrate that the model outperforms existing dense approaches.",
author = "Iaroslav Melekhov and Aleksei Tiulpin and Torsten Sattler and Marc Pollefeys and Esa Rahtu and Juho Kannala",
note = "jufoid=57596",
year = "2019",
month = "3",
day = "4",
doi = "10.1109/WACV.2019.00115",
language = "English",
series = "IEEE Winter Conference on Applications of Computer Vision",
publisher = "IEEE",
pages = "1034--1042",
booktitle = "2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - DGC-Net

T2 - Dense geometric correspondence network

AU - Melekhov, Iaroslav

AU - Tiulpin, Aleksei

AU - Sattler, Torsten

AU - Pollefeys, Marc

AU - Rahtu, Esa

AU - Kannala, Juho

N1 - jufoid=57596

PY - 2019/3/4

Y1 - 2019/3/4

N2 - This paper addresses the challenge of dense pixel correspondence estimation between two images. This problem is closely related to optical flow estimation task where Con-vNets (CNNs) have recently achieved significant progress. While optical flow methods produce very accurate results for the small pixel translation and limited appearance variation scenarios, they hardly deal with the strong geometric transformations that we consider in this work. In this paper, we propose a coarse-to-fine CNN-based framework that can leverage the advantages of optical flow approaches and extend them to the case of large transformations providing dense and subpixel accurate estimates. It is trained on synthetic transformations and demonstrates very good performance to unseen, realistic, data. Further, we apply our method to the problem of relative camera pose estimation and demonstrate that the model outperforms existing dense approaches.

AB - This paper addresses the challenge of dense pixel correspondence estimation between two images. This problem is closely related to optical flow estimation task where Con-vNets (CNNs) have recently achieved significant progress. While optical flow methods produce very accurate results for the small pixel translation and limited appearance variation scenarios, they hardly deal with the strong geometric transformations that we consider in this work. In this paper, we propose a coarse-to-fine CNN-based framework that can leverage the advantages of optical flow approaches and extend them to the case of large transformations providing dense and subpixel accurate estimates. It is trained on synthetic transformations and demonstrates very good performance to unseen, realistic, data. Further, we apply our method to the problem of relative camera pose estimation and demonstrate that the model outperforms existing dense approaches.

U2 - 10.1109/WACV.2019.00115

DO - 10.1109/WACV.2019.00115

M3 - Conference contribution

T3 - IEEE Winter Conference on Applications of Computer Vision

SP - 1034

EP - 1042

BT - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019

PB - IEEE

ER -