Device and method of detecting gradual shot transition in moving picture
    6.
    发明申请
    Device and method of detecting gradual shot transition in moving picture 审中-公开
    检测运动图像中渐变镜头转换的装置和方法

    公开(公告)号:US20070248243A1

    公开(公告)日:2007-10-25

    申请号:US11521318

    申请日:2006-09-15

    IPC分类号: G06K9/00 G06K9/46

    CPC分类号: G06K9/00711

    摘要: An apparatus for and a method of detecting a gradual shot transition in a moving picture The method includes: setting at least one feature point of a frame included in the moving picture, tracking the set at least one feature point, and setting a feature point of a subsequent frame, from the tracked at least one feature point; calculating a feature value corresponding to each feature point of each of the frames; and comparing corresponding feature values of corresponding feature points which are tracked for each of the frames, and determining that a gradual shot transition between the frames occurs when the comparison value is greater than a predetermined threshold value.

    摘要翻译: 一种用于检测运动图像中的渐变镜头转换的装置和方法。该方法包括:设置运动图像中包括的帧的至少一个特征点,跟踪所设置的至少一个特征点,以及设置特征点 跟踪的至少一个特征点的后续帧; 计算与每个帧的每个特征点对应的特征值; 以及比较对于每个帧跟踪的相应特征点的相应特征值,并且当所述比较值大于预定阈值时,确定所述帧之间的渐变镜头转换。

    Method and system for classifying scene for each person in video
    7.
    发明申请
    Method and system for classifying scene for each person in video 有权
    为视频中的每个人分类场景的方法和系统

    公开(公告)号:US20080166027A1

    公开(公告)日:2008-07-10

    申请号:US11882733

    申请日:2007-08-03

    IPC分类号: G06K9/00

    摘要: A method of classifying a scene for each person in a video, the method including: detecting a face within input video frames; detecting a shot change of the input video frames; extracting a person representation frame in the shot; performing a person clustering in the extracted person representation frame based on time information; detecting a scene change by separating a person portion from a background based on face extraction information, and comparing the person portion and the background; and merging similar clusters from the extracted person representation frame and performing a scene clustering for each person.

    摘要翻译: 一种对视频中的每个人分类场景的方法,所述方法包括:检测输入视频帧内的人脸; 检测输入视频帧的镜头变化; 在镜头中提取人物表示框架; 基于时间信息执行提取的人表示帧中的人聚类; 通过基于面部提取信息将人物部分与背景分开来检测场景变化,并且比较人物部分和背景; 并从提取的人表示框架合并类似的聚类,并为每个人执行场景聚类。

    Moving picture play system, method and medium thereof
    8.
    发明申请
    Moving picture play system, method and medium thereof 审中-公开
    运动图像播放系统,方法和介质

    公开(公告)号:US20080018739A1

    公开(公告)日:2008-01-24

    申请号:US11586605

    申请日:2006-10-26

    IPC分类号: H04N7/18

    摘要: A moving picture play system, method and medium which includes determining whether a driver of a vehicle is unable to view moving picture data, summarizing moving picture data when it is determined the driver of the vehicle is unable to view the moving picture data, and playing the summarized moving picture data or recorded moving picture data when it is determined the driver of the vehicle is able to view moving picture data.

    摘要翻译: 一种运动图像播放系统,方法和介质,其包括确定车辆的驾驶员是否不能观看运动图像数据,当确定车辆的驾驶员不能观看运动图像数据并且播放时,总结运动图像数据 当确定车辆的驾驶员能够观看运动图像数据时,总结的运动图像数据或记录的运动图像数据。

    Method and system for classifying scene for each person in video
    10.
    发明申请
    Method and system for classifying scene for each person in video 审中-公开
    为视频中的每个人分类场景的方法和系统

    公开(公告)号:US20120039515A1

    公开(公告)日:2012-02-16

    申请号:US13317509

    申请日:2011-10-20

    IPC分类号: G06K9/62

    摘要: Described is a method of classifying a scene for each person in a video, the method including: detecting a face within input video frames; detecting a shot change of the input video frames; extracting a person representation frame in the shot; performing a person clustering in the extracted person representation frame based on time information; detecting a scene change by separating a person portion from a background based on face extraction information, and comparing the person portion and the background; and merging similar clusters from the extracted person representation frame and performing a scene clustering for each person.

    摘要翻译: 描述了对视频中的每个人分类场景的方法,所述方法包括:检测输入视频帧内的人脸; 检测输入视频帧的镜头变化; 在镜头中提取人物表示框架; 基于时间信息执行提取的人表示帧中的人聚类; 通过基于面部提取信息将人物部分与背景分开来检测场景变化,并且比较人物部分和背景; 并从提取的人表示框架合并类似的聚类,并为每个人执行场景聚类。