TY - GEN
T1 - Cross-domain visual representations via unsupervised graph alignment
AU - Yang, Baoyao
AU - YUEN, Pong Chi
N1 - Funding Information:
This work was partially supported by the Science Faculty Research Grant of Hong Kong Baptist University, Hong Kong Research Grants Council General Research Fund: RGC/HKBU12200518.
PY - 2019/7/17
Y1 - 2019/7/17
N2 - In unsupervised domain adaptation, distributions of visual representations are mismatched across domains, which leads to the performance drop of a source model in the target domain. Therefore, distribution alignment methods have been proposed to explore cross-domain visual representations. However, most alignment methods have not considered the difference in distribution structures across domains, and the adaptation would subject to the insufficient aligned cross-domain representations. To avoid the misclassification/misidentification due to the difference in distribution structures, this paper proposes a novel unsupervised graph alignment method that aligns both data representations and distribution structures across the source and target domains. An adversarial network is developed for unsupervised graph alignment, which maps both source and target data to a feature space where data are distributed with unified structure criteria. Experimental results show that the graph-aligned visual representations achieve good performance on both cross-dataset recognition and cross-modal re-identification.
AB - In unsupervised domain adaptation, distributions of visual representations are mismatched across domains, which leads to the performance drop of a source model in the target domain. Therefore, distribution alignment methods have been proposed to explore cross-domain visual representations. However, most alignment methods have not considered the difference in distribution structures across domains, and the adaptation would subject to the insufficient aligned cross-domain representations. To avoid the misclassification/misidentification due to the difference in distribution structures, this paper proposes a novel unsupervised graph alignment method that aligns both data representations and distribution structures across the source and target domains. An adversarial network is developed for unsupervised graph alignment, which maps both source and target data to a feature space where data are distributed with unified structure criteria. Experimental results show that the graph-aligned visual representations achieve good performance on both cross-dataset recognition and cross-modal re-identification.
UR - http://www.scopus.com/inward/record.url?scp=85075428569&partnerID=8YFLogxK
U2 - 10.1609/aaai.v33i01.33015613
DO - 10.1609/aaai.v33i01.33015613
M3 - Conference contribution
AN - SCOPUS:85075428569
T3 - Proceedings of the AAAI Conference on Artificial Intelligence
SP - 5613
EP - 5620
BT - 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
PB - AAAI press
T2 - 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Annual Conference on Innovative Applications of Artificial Intelligence, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
Y2 - 27 January 2019 through 1 February 2019
ER -