中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

55 records (took 0.015 seconds)

    Automatic Temporal Layer Bit Allocation
    41.
    发明申请
    Automatic Temporal Layer Bit Allocation 有权
    自动时域分配

    公开(公告)号：US20120082237A1

    公开(公告)日：2012-04-05

    申请号：US12897365

    申请日：2010-10-04

    Applicant: Wonkap Jang ,  Michael Horowitz

    Inventor： Wonkap Jang ,  Michael Horowitz

    IPC: H04N7/26

    CPC classification number: H04N19/31 ,  H04N19/149 ,  H04N19/172 ,  H04N19/187 ,  H04N19/36

    Abstract: Disclosed herein are techniques and computer readable media containing instructions arranged to determine a bit allocation for a temporal base layer and at least one temporal enhancement layer. The techniques are enhanced to yield improved reproduced quality for the case where both a temporal base layer and at least one temporal enhancement layer is being used. An exemplary method for determining a bit allocation to one base layer includes initializing a bit allocation between the base layer and the at least one enhancement layer, determining a quality difference between the base layer and the at least one enhancement layer, expressed, for example, in the Quantizer Parameter of the last picture of each layer within a GOP, and determining a new bit allocation between the base layer and the at least one enhancement layer based on the quality difference and at least one constant factor.

    Abstract translation: 这里公开的技术和计算机可读介质包含被布置为确定时间基础层和至少一个时间增强层的比特分配的指令。 增强这些技术以在使用时基基层和至少一个时间增强层的情况下产生改进的再现质量。 用于确定对一个基本层的比特分配的示例性方法包括初始化基本层和至少一个增强层之间的比特分配，确定基本层和至少一个增强层之间的质量差异，例如， 在GOP的每个层的最后图像的量化器参数中，并且基于质量差异和至少一个常数因子确定基本层和至少一个增强层之间的新的比特分配。

    Delay Aware Rate Control In The Context Of Hierarchical P Picture Coding
    42.
    发明申请
    Delay Aware Rate Control In The Context Of Hierarchical P Picture Coding 有权
    在分层P图像编码的背景下的延迟感知速率控制

    公开(公告)号：US20120082214A1

    公开(公告)日：2012-04-05

    申请号：US13010570

    申请日：2011-01-20

    Applicant: Michael Horowitz ,  Wonkap Jang ,  Ofer Shapiro

    Inventor： Michael Horowitz ,  Wonkap Jang ,  Ofer Shapiro

    IPC: H04N7/26 ,  H04N7/32

    CPC classification number: H04N19/164 ,  H04N19/124 ,  H04N19/146 ,  H04N19/162 ,  H04N19/172

    Abstract: The present invention provides bit rate control techniques for a hybrid block based motion compensated predictive video encoder (e.g., H.264, as well as other video coding algorithms and standards). The techniques aim to reduce the picture-to-picture bit rate fluctuations within a prediction structure or over a short time interval that can lead to undesirable additional delay on links without over-provisioned capacity. In one embodiment of the present invention, the end to end delay is estimated. The end to end delay is used to determine limits on the modulation of the target coded picture size, with the goal of trading an acceptable amount of additional delay for a better coding efficiency.

    Abstract translation: 本发明提供了一种用于基于混合块的运动补偿预测视频编码器（例如，H.264以及其他视频编码算法和标准）的比特率控制技术。 这些技术旨在减少预测结构内的图像到图像比特率波动，或者在短时间间隔内，这可能导致链路上不期望的附加延迟而没有过度提供的容量。 在本发明的一个实施例中，估计端到端延迟。 端到端延迟用于确定目标编码图像大小的调制限制，目的是交换可接受的额外延迟量以获得更好的编码效率。

    Adaptive filter to improve H-264 video quality
    46.
    发明申请
    Adaptive filter to improve H-264 video quality 失效
    自适应滤波器提高H-264视频质量

    公开(公告)号：US20050025236A1

    公开(公告)日：2005-02-03

    申请号：US10633273

    申请日：2003-08-01

    Applicant: Yong Yan ,  Michael Horowitz ,  Athar Shah ,  Richard Hall

    Inventor： Yong Yan ,  Michael Horowitz ,  Athar Shah ,  Richard Hall

    IPC: H04N7/12 ,  H04N7/26

    CPC classification number: H04N19/86 ,  H04N19/80

    Abstract: A method for reducing or eliminating coding artifacts in video pictures processed using intra block prediction methods, for example, according to the H.264 standard. The causes of the coding artifacts are identified to be the intra prediction modes with prediction directions that are not in the direction of the raster scan. Filtering the affected blocks with a simple one-dimensional spatial filter will reduce or even eliminate these coding artifacts.

    Abstract translation: 一种减少或消除使用帧内预测方法处理的视频图像中的伪像的方法，例如根据H.264标准。 编码伪像的原因被识别为具有不在光栅扫描方向上的预测方向的帧内预测模式。 用简单的一维空间过滤器过滤受影响的块将减少甚至消除这些编码工件。

    Motion wake identification and control mechanism
    49.
    发明授权
    Motion wake identification and control mechanism 失效
    运动尾迹识别和控制机制

    公开(公告)号：US08780970B2

    公开(公告)日：2014-07-15

    申请号：US10328330

    申请日：2002-12-23

    Applicant: Michael Horowitz

    Inventor： Michael Horowitz

    IPC: H04B1/66

    CPC classification number: H04N7/148 ,  H04N19/107 ,  H04N19/124 ,  H04N19/137 ,  H04N19/139 ,  H04N19/14 ,  H04N19/176 ,  H04N19/503 ,  H04N19/51 ,  H04N19/61 ,  H04N19/70 ,  H04N19/89 ,  H04N19/895

    Abstract: The present invention provides a system and method for detecting macroblocks of a video frame that are located in one or more motion wake regions of the video frame. In one embodiment, a motion wake engine is configured to select a subset of macroblocks comprising non-peripherally located macroblocks selected to be inter-coded with motion vector magnitudes less than a predetermined motion vector threshold. Then, the motion wake engine identifies which macroblocks of the subset of macroblocks are located in first, second, third, or fourth motion wake regions of the video frame based upon a motion vector analysis of nearest neighbor macroblocks. In another embodiment of the invention, the motion wake engine applies error concealment/video improvement techniques to the identified macroblocks for video error concealment and for reducing visual artifacts associated with the motion wake regions.

    Abstract translation: 本发明提供一种用于检测位于视频帧的一个或多个运动唤醒区域中的视频帧的宏块的系统和方法。 在一个实施例中，运动唤醒引擎被配置为选择包括被选择为与小于预定运动向量阈值的运动矢量量值进行互编码的非外围定位的宏块的宏块子集。 然后，运动唤醒引擎基于最近相邻宏块的运动矢量分析来识别宏块子集中的哪个宏块位于视频帧的第一，第二，第三或第四运动唤醒区域中。 在本发明的另一个实施例中，运动唤醒引擎将错误隐藏/视频改进技术应用于所识别的宏块用于视频错误隐藏和用于减少与运动唤醒区域相关联的视觉伪像。

Patent Agency Ranking