【24h】

Learning-based face tracking and pose estimation in unconstrained environments

机译:在不受限制的环境中基于学习的面部跟踪和姿势估计

获取原文
获取原文并翻译 | 示例
           

摘要

Computers are getting widely used as personal and mobile equipment, the environment where they are used is unconstrained and thus much difficult to predict. As a prospective HCI device, human face has been well studied in constrained environment. In this report, we propose an interactive color modeling method that is easily adapted to the unconstrained environment. Also, we propose to use the self-organizing map (SOM) for face pose estimation. Experiments using image sequences taken from outdoors and indoors have been done, the preliminary result is encouraging.
机译:计算机正被广泛用作个人和移动设备,它们的使用环境不受限制,因此很难预测。作为一种预期的HCI设备,人脸在受限的环境中得到了很好的研究。在此报告中,我们提出了一种交互式颜色建模方法,该方法易于适应不受约束的环境。另外,我们建议使用自组织映射(SOM)进行脸部姿势估计。已经进行了使用从室外和室内获取的图像序列的实验,初步结果令人鼓舞。

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