TY - GEN
T1 - Feature constrained by pixel
T2 - 26th ACM Multimedia conference, MM 2018
AU - Shao, Rui
AU - LAN, Xiangyuan
AU - YUEN, Pong Chi
N1 - Funding Information:
The authors would like to thank Dr. Jiawei Li and Dr. Baoyao Yang for their helpful suggestions. This project is partially supported by Hong Kong RGC General Research Fund HKBU 12200518.
PY - 2018/10/15
Y1 - 2018/10/15
N2 - In multimedia analysis, one objective of unsupervised visual domain adaptation is to train a classifier that works well on a target domain given labeled source samples and unlabeled target samples. Feature alignment of two domains is the key issue which should be addressed to achieve this objective. Inspired by the recent study of Generative Adversarial Networks (GAN) in domain adaptation, this paper proposes a new model based on Generative Adversarial Network, named Hierarchical Adversarial Deep Network (HADN), which jointly optimizes the feature-level and pixel-level adversarial adaptation within a hierarchical network structure. Specifically, the hierarchical network structure ensures that the knowledge from pixel-level adversarial adaptation can be back propagated to facilitate the feature-level adaptation, which achieves a better feature alignment under the constraint of pixel-level adversarial adaptation. Extensive experiments on various visual recognition tasks show that the proposed method performs favorably against or better than competitive state-of-the-art methods.
AB - In multimedia analysis, one objective of unsupervised visual domain adaptation is to train a classifier that works well on a target domain given labeled source samples and unlabeled target samples. Feature alignment of two domains is the key issue which should be addressed to achieve this objective. Inspired by the recent study of Generative Adversarial Networks (GAN) in domain adaptation, this paper proposes a new model based on Generative Adversarial Network, named Hierarchical Adversarial Deep Network (HADN), which jointly optimizes the feature-level and pixel-level adversarial adaptation within a hierarchical network structure. Specifically, the hierarchical network structure ensures that the knowledge from pixel-level adversarial adaptation can be back propagated to facilitate the feature-level adaptation, which achieves a better feature alignment under the constraint of pixel-level adversarial adaptation. Extensive experiments on various visual recognition tasks show that the proposed method performs favorably against or better than competitive state-of-the-art methods.
KW - Deep learning
KW - Generative Adversarial Network
KW - Unsupervised domain adaptation
UR - http://www.scopus.com/inward/record.url?scp=85058242992&partnerID=8YFLogxK
U2 - 10.1145/3240508.3240562
DO - 10.1145/3240508.3240562
M3 - Conference contribution
AN - SCOPUS:85058242992
T3 - MM 2018 - Proceedings of the 2018 ACM Multimedia Conference
SP - 220
EP - 228
BT - MM 2018 - Proceedings of the 2018 ACM Multimedia Conference
PB - Association for Computing Machinery (ACM)
Y2 - 22 October 2018 through 26 October 2018
ER -