arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1707.09798cs.CV
收藏

基于可重构生成对抗网络的无监督视觉属性传递

Unsupervised Visual Attribute Transfer with Reconfigurable Generative Adversarial Networks

SK T-Brain

Taeksoo Kim, Byoungjip Kim, Moonsu Cha, Jiwon Kim

摘要:

学习传递视觉属性需要监督数据集。学习传递函数需要具有相同身份的不同属性值的对应图像。这在很大程度上限制了它们的应用,因为捕获它们通常是一项艰巨的任务。为了解决这个问题,我们提出了一个无监督的方法来学习转移视觉属性。所提出的方法可以学习的传递函数没有任何对应的图像。通过对多个无监督属性迁移任务的可视化结果进行检验,验证了该方法的有效性。

英文摘要:

Learning to transfer visual attributes requires supervision dataset. Corresponding images with varying attribute values with the same identity are required for learning the transfer function. This largely limits their applications, because capturing them is often a difficult task. To address the issue, we propose an unsupervised method to learn to transfer visual attribute. The proposed method can learn the transfer function without any corresponding images. Inspecting visualization results from various unsupervised attribute transfer tasks, we verify the effectiveness of the proposed method.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑