首页> 外文期刊>電子情報通信学会技術研究報告. 音声. Speech >Automatic face tracking in video sequence with zoom change and face image synthesis by match-moving
【24h】

Automatic face tracking in video sequence with zoom change and face image synthesis by match-moving

机译:视频序列中的自动人脸跟踪,具有缩放更改和通过匹配移动进行人脸图像合成

获取原文
获取原文并翻译 | 示例
           

摘要

In a special effect of movie creation, a manual match-moving can realize the replacement of the face in film with movie star, however, this process need a creator's experience and long time working. Also in stand-in system of foreign movie, lip synchronization is not achieved and sometimes spoken work is limited by mouth movement. In this paper, the location and angle of face are automatically estimated in video sequence and full face expression or mouth movement are replaced with new one. 3D face template is introduced to realize an accurate face tracking to realize automatic match-moving and stand-in system.
机译:在电影创作的特殊效果中,通过手动配对可以实现用电影明星替换电影中的面孔,但是,此过程需要创作者的经验和长时间的工作。同样在外国电影的站立式系统中,也无法实现口型同步,有时口头动作会限制口头工作。在本文中,视频序列会自动估计脸部位置和角度,并用新脸部表情或嘴巴动作代替脸部表情。引入了3D人脸模板以实现精确的人脸跟踪,从而实现自动的移动匹配和备用系统。

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