Abstract
In this paper, we investigate a principle way to learn a common feature space for data of different modalities (e.g. image and text), so that the simi......
小提示：本篇文献需要登录阅读全文，点击跳转登录
Full Text Link
