TY - GEN
T1 - Iterative Image Tfor Unsupervised Domain Adaptation
AU - Chhabra, Sachin
AU - Venkateswara, Hemanth
AU - Li, Baoxin
N1 - Publisher Copyright:
© 2021 ACM.
PY - 2021/10/24
Y1 - 2021/10/24
N2 - In this paper, we propose an image-Translation-based unsupervised domain adaptation approach that iteratively trains an image translation and a classification network using each other. In Phase A, a classification network is used to guide the image translation to preserve the content and generate images. In Phase B, the generated images are used to train the classification network. With each step, the classification network and generator improve each other to learn the target domain representation. Detailed analysis and the experiments are testimony of the strength of our approach.
AB - In this paper, we propose an image-Translation-based unsupervised domain adaptation approach that iteratively trains an image translation and a classification network using each other. In Phase A, a classification network is used to guide the image translation to preserve the content and generate images. In Phase B, the generated images are used to train the classification network. With each step, the classification network and generator improve each other to learn the target domain representation. Detailed analysis and the experiments are testimony of the strength of our approach.
KW - generative domain adaptation
KW - image translation
KW - source-like target
KW - ternary feature alignment
KW - unsupervised domain adaptation
UR - http://www.scopus.com/inward/record.url?scp=85119091561&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85119091561&partnerID=8YFLogxK
U2 - 10.1145/3476098.3485050
DO - 10.1145/3476098.3485050
M3 - Conference contribution
AN - SCOPUS:85119091561
T3 - MULL 2021 - Proceedings of the 1st Workshop on Multimedia Understanding with Less Labeling, co-located with ACM MM 2021
SP - 37
EP - 44
BT - MULL 2021 - Proceedings of the 1st Workshop on Multimedia Understanding with Less Labeling, co-located with ACM MM 2021
PB - Association for Computing Machinery, Inc
T2 - 1st Workshop on Multimedia Understanding with Less Labeling, MULL 2021, co-located with ACM MM 2021
Y2 - 24 October 2021
ER -