公开(公告)号：WO2010071442A1

公开(公告)日：2010-06-24

申请号：PCT/NO2009/000408

申请日：2009-11-27

Applicant: TANDBERG TELECOM AS ,  CATCHPOLE, Jason ,  COCKERTON, Craig

Inventor： CATCHPOLE, Jason ,  COCKERTON, Craig

IPC: H04N7/15 ,  G06K9/62

CPC classification number: G06K9/00288 ,  G06F17/30793 ,  G06F17/30817 ,  G06K2009/00328

Abstract: The present invention provides a system and a method making names of participants in an archived conference or presentation searchable after being stored in the archive server. According to the invention, one or more media streams coded according to H.323 or SIP are transmitted to a streaming/archiving server for converting multimedia content into a standard streaming format. A face recognition unit analyzes the video part of the multimedia content to detect faces and identify the detected faces. The face recognition unit utilizes protocol information and other video conferencing recourses to greatly reduce the search space during face recognition process, hence reducing computational cost and improving robustness of recognitions.

Abstract translation: 本发明提供一种系统和方法，其在被存档在存档服务器中之后，可以在可归档的会议或表示中创建参与者的名称。 根据本发明，根据H.323或SIP编码的一个或多个媒体流被传送到用于将多媒体内容转换为标准流格式的流/归档服务器。 面部识别单元分析多媒体内容的视频部分以检测面部并识别检测到的面部。 面部识别单元利用协议信息和其他视频会议资源，大大减少了人脸识别过程中的搜索空间，从而降低了计算成本，提高了识别的鲁棒性。

公开(公告)号：EP2380349A1

公开(公告)日：2011-10-26

申请号：EP09833677.9

申请日：2009-11-27

Applicant: Tandberg Telecom AS

Inventor： CATCHPOLE, Jason ,  COCKERTON, Craig

IPC: H04N7/15 ,  G06K9/62

CPC classification number: G06K9/00288 ,  G06F17/30793 ,  G06F17/30817 ,  G06K2009/00328

Abstract: The present invention provides a system and a method making names of participants in an archived conference or presentation searchable after being stored in the archive server. According to the invention, one or more media streams coded according to H.323 or SIP are transmitted to a streaming/archiving server for converting multimedia content into a standard streaming format. A face recognition unit analyzes the video part of the multimedia content to detect faces and identify the detected faces. The face recognition unit utilizes protocol information and other video conferencing recourses to greatly reduce the search space during face recognition process, hence reducing computational cost and improving robustness of recognitions.