专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US06670963B2 Visual attention model 有权
公开(公告)号：US06670963B2
公开(公告)日：2003-12-30
申请号：US09764726
申请日：2001-01-17
申请人： Wilfried M. Osberger
发明人： Wilfried M. Osberger
IPC分类号： G09G500
CPC分类号： G06K9/00711 , G06K9/036 , G06K9/3233 , G06T7/11 , G06T7/215 , G06T2207/10016 , H04N19/186 , H04N19/503 , H04N19/527 , H04N19/533
摘要： An improved visual attention model uses a robust adaptive segmentation algorithm to divide a current frame of a video sequence into a plurality of regions based upon both color and luminance, with each region being processed in parallel by a plurality of spatial feature algorithms including color and skin to produce respective spatial importance maps. The current frame and a previous frame are also processed to produce motion vectors for each block of the current frame, the motion vectors being compensated for camera motion, and the compensated motion vectors being converted to produce a temporal importance map. The spatial and temporal importance maps are combined using weighting based upon eye movement studies.

2. 发明授权

US07099518B2 Measurement of blurring in video sequences 有权
标题翻译：视频序列模糊测量
公开(公告)号：US07099518B2
公开(公告)日：2006-08-29
申请号：US10198944
申请日：2002-07-18
申请人： Bei Li , Wilfried M. Osberger
发明人： Bei Li , Wilfried M. Osberger
IPC分类号： G06K9/40 , G06K9/48
CPC分类号： H04N17/00 , H04N5/208 , H04N5/21 , H04N17/004
摘要： A method for determine blurring in a test video sequence due to video processing includes detecting blocks within each frame of the test video sequence that have valid image edges. An edge point within each valid image edge block is selected and a series of points defining an edge profile in the block along a line normal to the valid image edge at each edge point is defined from an enhanced edge map in which video processing blockiness artifacts have been removed. From the edge profile a blurring value is estimated for each frame or group of frames within the test video sequence. Additionally a reference blurring value may be derived from a reference video sequence corresponding to the test video sequence, which reference blurring value may be generated at a source of the reference video sequence and transmitted with the test video sequence to a receiver or may be generated at the receiver. The reference blurring value is then compared with the blurring value from the test video sequence to produce a relative blurring value for the test video sequence.
摘要翻译：由于视频处理，用于确定测试视频序列中的模糊的方法包括检测具有有效图像边缘的测试视频序列的每帧内的块。选择每个有效图像边缘块内的边缘点，并且沿着沿着垂直于每个边缘点处的有效图像边缘的线的块在块中定义边缘轮廓的一系列点从其中视频处理块状伪影具有的增强边缘图定义已被删除从边缘轮廓中，针对测试视频序列内的每个帧或一组帧估计模糊值。另外，可以从对应于测试视频序列的参考视频序列导出参考模糊值，该参考模糊值可以在参考视频序列的源处生成并且与测试视频序列一起发送到接收机，或者可以在收件人。然后将参考模糊值与来自测试视频序列的模糊值进行比较，以产生测试视频序列的相对模糊值。

3. 发明授权

US07586515B2 Instrument for real-time video quality measurement 有权
标题翻译：仪器用于实时视频质量测量
公开(公告)号：US07586515B2
公开(公告)日：2009-09-08
申请号：US11135726
申请日：2005-05-23
申请人： Gale L. Straney , Wilfried M. Osberger
发明人： Gale L. Straney , Wilfried M. Osberger
IPC分类号： H04N17/00 , H04N17/02
CPC分类号： H04N17/00
摘要： A realtime video quality measurement instrument may be configured for both double-ended and single-ended operation. For double-ended operation reference and test video signals are stored in respective buffers and spatial/temporally aligned. Desired quality measurements are performed on the aligned frames of the test and reference video signals according to stored setup instructions. For single-ended operation the reference video signal and a signature for the reference video signal are pre-stored together with the desired quality measurements for the frames of the reference video signal. Then the test video signal is received, signatures determined, and the test and reference video signals aligned using the signatures. The desired quality measurements are then performed on the aligned frames of the test and reference video signals.
摘要翻译：实时视频质量测量仪器可以配置为双端和单端操作。对于双端操作，参考和测试视频信号存储在相应的缓冲器中并进行空间/时间对准。根据存储的设置指令，对测试和参考视频信号的对齐帧执行所需的质量测量。对于单端操作，参考视频信号和参考视频信号的签名与参考视频信号的帧的期望质量测量一起被预先存储。然后接收测试视频信号，确定签名，并使用签名对齐测试和参考视频信号。然后对测试和参考视频信号的对准帧执行所需的质量测量。

4. 发明授权

US06738099B2 Robust camera motion estimation for video sequences 有权
标题翻译：视频序列的强大的相机运动估计
公开(公告)号：US06738099B2
公开(公告)日：2004-05-18
申请号：US09788135
申请日：2001-02-16
申请人： Wilfried M. Osberger
发明人： Wilfried M. Osberger
IPC分类号： H04N712
CPC分类号： H04N5/145
摘要： A robust technique for estimating camera motion parameters calculates the motion vectors for a current frame vis a vis a previous frame using a multi-scale block matching technique. The means of the motion vectors for the current and the previous frames are compared for a temporal discontinuity, the detection of such temporal discontinuity as a temporal repetition, such as frozen field or 3:2 pulldown, terminating the processing of the current frame with the camera motion parameter estimate for the previous frame being used for the current frame. Otherwise motion vectors for spatially flat areas and text/graphic overlay areas are discarded and an error-of-fit for the previous frame is tested to determine initial parameters for an iterative camera motion estimation process. If the error-of-fit is less than a predetermined threshold, then the camera motion parameters for the previous frame are used as the initial parameters, otherwise a best least squares fit is used as the initial parameters. Outlier motion vectors are removed and the camera motion parameters are recalculated using a least squares best fit in an iterative process until either the error-of-fit for the current frame is less than the predetermined threshold or the number of iterations exceeds a maximum. The estimated camera motion parameters and associated error-of-fit are provided as an output.
摘要翻译：用于估计摄像机运动参数的鲁棒技术使用多尺度块匹配技术来计算当前帧相对于先前帧的运动矢量。对于当前帧和前一帧的运动矢量的平均值进行时间不连续性的检测，如时间重复（例如，冻结场或3：2下拉）等时间不连续性的检测，终止当前帧的处理用于当前帧的前一帧的相机运动参数估计。否则，丢弃空间平坦区域和文本/图形覆盖区域的运动矢量，并测试前一帧的拟合误差，以确定迭代相机运动估计过程的初始参数。如果拟合误差小于预定阈值，则将先前帧的相机运动参数用作初始参数，否则将使用最佳最小二乘拟合作为初始参数。去除异常运动矢量，并使用迭代过程中的最小二乘法最佳拟合重新计算摄像机运动参数，直到当前帧的拟合误差小于预定阈值或迭代次数超过最大值。提供估计的摄像机运动参数和相关的拟合误差作为输出。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式