• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Conference on Multimedia Modeling: Modeling Multimedia Information and Systems (MMM 2000) Nov 13-15, 2000, Nagano, Japan >AN AUTOMATED COMPRESSED-DOMAIN FACE DETECTION METHOD FOR VIDEO STRATIFICATION
【24h】

AN AUTOMATED COMPRESSED-DOMAIN FACE DETECTION METHOD FOR VIDEO STRATIFICATION

机译：用于视频分层的自动压缩域面部检测方法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A news video can be modeled using the stratification approach by identifying, among other entities, human faces appearing in the video stream. To facilitate this, we need to develop techniques to detect and track human faces in video. This paper presents a frontal face detection method that uses the gradient energy representation extracted directly from the MPEG video. The gradient energy representation permits pertinent facial features of high contrast, such as eyes, nose and mouth, to be highlighted. The method encodes the local appearance and layout of these pertinent facial features in terms of a set of rules. The parameters for the rules are learnt from sample faces. To detect faces, candidate video frames are processed in two steps. First, we use the rule-based model to locate potential face patterns within a video frame at multiple scales and locations. Second, we perform skin-color verification to eliminate falsely detected regions. The main contribution of this work is in developing an efficient scale and position invariant method to detect faces that operates in a transformed gradient energy space in compressed domain. The system is tested on selected video clips from MPEG-7 data set and was found to be effective.
机译：可以使用分层方法对新闻视频进行建模，方法是识别视频流中出现的人脸以及其他实体。为此，我们需要开发技术来检测和跟踪视频中的人脸。本文提出了一种正面人脸检测方法，该方法使用直接从MPEG视频中提取的梯度能量表示。梯度能量表示可以突出显示高对比度的相关面部特征，例如眼睛，鼻子和嘴巴。该方法根据一组规则对这些相关面部特征的局部外观和布局进行编码。规则的参数是从样本面上学习的。为了检测面部，分两个步骤处理候选视频帧。首先，我们使用基于规则的模型来定位视频帧内多个比例和位置的潜在面部模式。其次，我们执行肤色验证以消除错误检测的区域。这项工作的主要贡献在于开发了一种有效的比例尺和位置不变方法，以检测在压缩域中变换的梯度能量空间中工作的人脸。在从MPEG-7数据集中选择的视频片段上对该系统进行了测试，发现该系统有效。

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