Fu Huan, Gong Mingming, Wang Chaohui, Batmanghelich Kayhan, Zhang Kun, Tao Dacheng
UBTECH Sydney AI Centre, School of Computer Science, FEIT, University of Sydney, Darlington, NSW 2008, Australia.
Department of Biomedical Informatics, University of Pittsburgh.
Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2019 Jun;2019:2422-2431. doi: 10.1109/cvpr.2019.00253. Epub 2020 Jan 9.
Unsupervised domain mapping aims to learn a function G to translate domain to in the absence of paired examples. Finding the optimal without paired data is an ill-posed problem, so appropriate constraints are required to obtain reasonable solutions. While some prominent constraints such as cycle consistency and distance preservation successfully constrain the solution space, they overlook the special properties of images that simple geometric transformations do not change the image's semantic structure. Based on this special property, we develop a geometry-consistent generative adversarial network (), which enables one-sided unsupervised domain mapping. takes the original image and its counterpart image transformed by a predefined geometric transformation as inputs and generates two images in the new domain coupled with the corresponding geometry-consistency constraint. The geometry-consistency constraint reduces the space of possible solutions while keep the correct solutions in the search space. Quantitative and qualitative comparisons with the baseline () and the state-of-the-art methods including [66] and [5] demonstrate the effectiveness of our method.
无监督域映射旨在学习一个函数G，以便在没有配对示例的情况下将一个域转换为另一个域。在没有配对数据的情况下找到最优的G是一个不适定问题，因此需要适当的约束来获得合理的解决方案。虽然一些突出的约束，如循环一致性和距离保持，成功地限制了解决方案空间，但它们忽略了图像的特殊属性，即简单的几何变换不会改变图像的语义结构。基于这一特殊属性，我们开发了一种几何一致的生成对抗网络（），它能够实现单边无监督域映射。该网络将原始图像及其通过预定义几何变换转换后的对应图像作为输入，并在新域中生成两个图像，同时施加相应的几何一致性约束。几何一致性约束减少了可能的解决方案空间，同时在搜索空间中保留了正确的解决方案。与基线（）以及包括[66]和[5]在内的最新方法进行的定量和定性比较证明了我们方法的有效性。
