Next Article in Journal
An Experimental Test Proposal to Study Human Behaviour in Fires Using Virtual Environments
Previous Article in Journal
High-Sensitivity Goos-Hänchen Shifts Sensor Based on BlueP-TMDCs-Graphene Heterostructure
Open AccessArticle

C2DAN: An Improved Deep Adaptation Network with Domain Confusion and Classifier Adaptation

by Han Sun 1,2,*, Xinyi Chen 1,2, Ling Wang 1,2, Dong Liang 1,2, Ningzhong Liu 1,2 and Huiyu Zhou 3
1
College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
2
MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Nanjing 211106, China
3
School of Informatics, University of Leicester, Leicester LE1 7RH, UK
*
Author to whom correspondence should be addressed.
Sensors 2020, 20(12), 3606; https://doi.org/10.3390/s20123606
Received: 3 June 2020 / Revised: 20 June 2020 / Accepted: 23 June 2020 / Published: 26 June 2020
(This article belongs to the Section Sensor Networks)
Deep neural networks have been successfully applied in domain adaptation which uses the labeled data of source domain to supplement useful information for target domain. Deep Adaptation Network (DAN) is one of these efficient frameworks, it utilizes Multi-Kernel Maximum Mean Discrepancy (MK-MMD) to align the feature distribution in a reproducing kernel Hilbert space. However, DAN does not perform very well in feature level transfer, and the assumption that source and target domain share classifiers is too strict in different adaptation scenarios. In this paper, we further improve the adaptability of DAN by incorporating Domain Confusion (DC) and Classifier Adaptation (CA). To achieve this, we propose a novel domain adaptation method named C2DAN. Our approach first enables Domain Confusion (DC) by using a domain discriminator for adversarial training. For Classifier Adaptation (CA), a residual block is added to the source domain classifier in order to learn the difference between source classifier and target classifier. Beyond validating our framework on the standard domain adaptation dataset office-31, we also introduce and evaluate on the Comprehensive Cars (CompCars) dataset, and the experiment results demonstrate the effectiveness of the proposed framework C2DAN. View Full-Text
Keywords: transfer learning; domain adaptation; MK-MMD; domain confusion; classifier adaptation; vehicle classification transfer learning; domain adaptation; MK-MMD; domain confusion; classifier adaptation; vehicle classification
Show Figures

Graphical abstract

MDPI and ACS Style

Sun, H.; Chen, X.; Wang, L.; Liang, D.; Liu, N.; Zhou, H. C2DAN: An Improved Deep Adaptation Network with Domain Confusion and Classifier Adaptation. Sensors 2020, 20, 3606.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop