【24h】

Facial Morphe via Domain Translation and FM~2RLS

机译:面部变形通过域翻译和fm〜2rls

获取原文

摘要

In this paper, we present a novel approach for face morphe. We use a recent domain transfer technique to generatetarget expression and combine a robust image deformation technique to obtain high realistic facial morphe. Weaddressed the facial mophe problem through three steps. Firstly, Domain transfer technique is introduced totransfer an image from one domain to another. Secondly, use a face alignment algorithm to locate accuratefacial landmark points for both domain transferred face and target face, then align them with a global similaritytransformation to eliminate their inconsistency in pose, size and position. Then we employ the FM~2RLS methodto deform the domain transferred face into the target image, let the images pairs align in pixel level. To validatethe effectiveness of the proposed approach, extensive experiments on real images shows accurate results of ourmethod, which is superior to the current state-of-the-art methods.
机译:在本文中,我们提出了一种面部变形的新方法。我们使用最近的域传输技术生成目标表达并结合鲁棒图像变形技术,获得高现实面部变形。我们通过三个步骤解决了面部Mophe问题。首先,介绍了域转移技术将图像从一个域转移到另一个域。其次,使用面向对准算法来定位准确两个域转移面部和目标面部的面部地标点,然后将它们与全局相似性对齐转型以消除姿势,大小和位置的不一致。然后我们使用FM〜2rls方法要将域转移到目标图像中的变形,使图像对在像素级别对齐。验证拟议方法的有效性,对真实形象的大量实验显示了我们的准确结果方法,其优于当前最先进的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号