TY - GEN
T1 - Learning Multi-view Generator Network for Shared Representation
AU - Han, Tian
AU - Xing, Xianglei
AU - Wu, Ying Nian
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/11/26
Y1 - 2018/11/26
N2 - Multi-view representation learning is challenging because different views contain both the common structure and the complex view specific information. The traditional generative models may not be effective in such situation, since view-specific and common information cannot be well separated, which may cause problems for downstream vision tasks. In this paper, we introduce a multi-view generator model to solve the problem of multi-view generation and recognition in a unified framework. We propose a multi-view alternating back-propagation algorithm to learn multi-view generator networks by allowing them to share common latent factors. Our experiments show that the proposed method is effective for both image generation and recognition. Specifically, we first qualitatively demonstrate that our model can rotate and complete faces accurately. Then we show that our model can achieve state-of-art or competitive recognition performances through quantitative comparisons.
AB - Multi-view representation learning is challenging because different views contain both the common structure and the complex view specific information. The traditional generative models may not be effective in such situation, since view-specific and common information cannot be well separated, which may cause problems for downstream vision tasks. In this paper, we introduce a multi-view generator model to solve the problem of multi-view generation and recognition in a unified framework. We propose a multi-view alternating back-propagation algorithm to learn multi-view generator networks by allowing them to share common latent factors. Our experiments show that the proposed method is effective for both image generation and recognition. Specifically, we first qualitatively demonstrate that our model can rotate and complete faces accurately. Then we show that our model can achieve state-of-art or competitive recognition performances through quantitative comparisons.
KW - Gait recognition
KW - Generator networks
KW - Multi-view learning
UR - http://www.scopus.com/inward/record.url?scp=85059755782&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059755782&partnerID=8YFLogxK
U2 - 10.1109/ICPR.2018.8545421
DO - 10.1109/ICPR.2018.8545421
M3 - Conference contribution
AN - SCOPUS:85059755782
T3 - Proceedings - International Conference on Pattern Recognition
SP - 2062
EP - 2068
BT - 2018 24th International Conference on Pattern Recognition, ICPR 2018
T2 - 24th International Conference on Pattern Recognition, ICPR 2018
Y2 - 20 August 2018 through 24 August 2018
ER -