A domain agnostic normalization layer for unsupervised adversarial domain adaptation

R.R.F.M. Romijnders, Panagiotis Meletis, Gijs Dubbelman

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

Uittreksel

We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We show that conventional normalization layers worsen the performance of current Unsupervised Adversarial Domain Adaption (UADA), which is a method to improve network performance on unlabeled data sets and the focus of our research. Therefore, we propose a novel Domain Agnostic Normalization layer and thereby unlock the benefits of normalization layers for unsupervised adversarial domain adaptation. In our evaluation, we adapt from the synthetic GTA5 data set to the real Cityscapes data set, a common benchmark experiment, and surpass the state-of-the-art. As our normalization layer is domain agnostic at test time, we furthermore demonstrate that UADA using Domain Agnostic Normalization improves performance on unseen domains, specifically on Apolloscape and Mapillary.
Originele taal-2Engels
Titel2019 IEEE Winter Conference on Applications of Computer Vision (WACV)
Plaats van productiePiscataway
UitgeverijInstitute of Electrical and Electronics Engineers
Pagina's1866-1875
Aantal pagina's10
ISBN van elektronische versie978-1-7281-1975-5
DOI's
StatusGepubliceerd - 4 mrt 2019
Evenement2019 IEEE Winter Conference on Applications of Computer Vision (WACV) - Waikoloa Village, Verenigde Staten van Amerika
Duur: 7 jan 201911 jan 2019

Publicatie series

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

Congres

Congres2019 IEEE Winter Conference on Applications of Computer Vision (WACV)
LandVerenigde Staten van Amerika
StadWaikoloa Village
Periode7/01/1911/01/19

Vingerafdruk

Network performance
Semantics
Neural networks
Experiments

Citeer dit

Romijnders, R. R. F. M., Meletis, P., & Dubbelman, G. (2019). A domain agnostic normalization layer for unsupervised adversarial domain adaptation. In 2019 IEEE Winter Conference on Applications of Computer Vision (WACV) (blz. 1866-1875). [8658995] (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019). Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/WACV.2019.00203
Romijnders, R.R.F.M. ; Meletis, Panagiotis ; Dubbelman, Gijs. / A domain agnostic normalization layer for unsupervised adversarial domain adaptation. 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). Piscataway : Institute of Electrical and Electronics Engineers, 2019. blz. 1866-1875 (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019).
@inproceedings{16c8ca1f454b42c4995e4176b3aabb49,
title = "A domain agnostic normalization layer for unsupervised adversarial domain adaptation",
abstract = "We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We show that conventional normalization layers worsen the performance of current Unsupervised Adversarial Domain Adaption (UADA), which is a method to improve network performance on unlabeled data sets and the focus of our research. Therefore, we propose a novel Domain Agnostic Normalization layer and thereby unlock the benefits of normalization layers for unsupervised adversarial domain adaptation. In our evaluation, we adapt from the synthetic GTA5 data set to the real Cityscapes data set, a common benchmark experiment, and surpass the state-of-the-art. As our normalization layer is domain agnostic at test time, we furthermore demonstrate that UADA using Domain Agnostic Normalization improves performance on unseen domains, specifically on Apolloscape and Mapillary.",
author = "R.R.F.M. Romijnders and Panagiotis Meletis and Gijs Dubbelman",
year = "2019",
month = "3",
day = "4",
doi = "10.1109/WACV.2019.00203",
language = "English",
series = "Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019",
publisher = "Institute of Electrical and Electronics Engineers",
pages = "1866--1875",
booktitle = "2019 IEEE Winter Conference on Applications of Computer Vision (WACV)",
address = "United States",

}

Romijnders, RRFM, Meletis, P & Dubbelman, G 2019, A domain agnostic normalization layer for unsupervised adversarial domain adaptation. in 2019 IEEE Winter Conference on Applications of Computer Vision (WACV)., 8658995, Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019, Institute of Electrical and Electronics Engineers, Piscataway, blz. 1866-1875, Waikoloa Village, Verenigde Staten van Amerika, 7/01/19. https://doi.org/10.1109/WACV.2019.00203

A domain agnostic normalization layer for unsupervised adversarial domain adaptation. / Romijnders, R.R.F.M.; Meletis, Panagiotis; Dubbelman, Gijs.

2019 IEEE Winter Conference on Applications of Computer Vision (WACV). Piscataway : Institute of Electrical and Electronics Engineers, 2019. blz. 1866-1875 8658995 (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019).

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

TY - GEN

T1 - A domain agnostic normalization layer for unsupervised adversarial domain adaptation

AU - Romijnders, R.R.F.M.

AU - Meletis, Panagiotis

AU - Dubbelman, Gijs

PY - 2019/3/4

Y1 - 2019/3/4

N2 - We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We show that conventional normalization layers worsen the performance of current Unsupervised Adversarial Domain Adaption (UADA), which is a method to improve network performance on unlabeled data sets and the focus of our research. Therefore, we propose a novel Domain Agnostic Normalization layer and thereby unlock the benefits of normalization layers for unsupervised adversarial domain adaptation. In our evaluation, we adapt from the synthetic GTA5 data set to the real Cityscapes data set, a common benchmark experiment, and surpass the state-of-the-art. As our normalization layer is domain agnostic at test time, we furthermore demonstrate that UADA using Domain Agnostic Normalization improves performance on unseen domains, specifically on Apolloscape and Mapillary.

AB - We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We show that conventional normalization layers worsen the performance of current Unsupervised Adversarial Domain Adaption (UADA), which is a method to improve network performance on unlabeled data sets and the focus of our research. Therefore, we propose a novel Domain Agnostic Normalization layer and thereby unlock the benefits of normalization layers for unsupervised adversarial domain adaptation. In our evaluation, we adapt from the synthetic GTA5 data set to the real Cityscapes data set, a common benchmark experiment, and surpass the state-of-the-art. As our normalization layer is domain agnostic at test time, we furthermore demonstrate that UADA using Domain Agnostic Normalization improves performance on unseen domains, specifically on Apolloscape and Mapillary.

UR - http://www.scopus.com/inward/record.url?scp=85063592212&partnerID=8YFLogxK

U2 - 10.1109/WACV.2019.00203

DO - 10.1109/WACV.2019.00203

M3 - Conference contribution

T3 - Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019

SP - 1866

EP - 1875

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

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Romijnders RRFM, Meletis P, Dubbelman G. A domain agnostic normalization layer for unsupervised adversarial domain adaptation. In 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). Piscataway: Institute of Electrical and Electronics Engineers. 2019. blz. 1866-1875. 8658995. (Proceedings - 2019 IEEE Winter Conference on Applications of Computer Vision, WACV 2019). https://doi.org/10.1109/WACV.2019.00203