Learning Multi-view Generator Network for Shared Representation

Tian Han, Xianglei Xing, Ying Nian Wu:

Abstract: 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-specificand 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.