公开(公告)号：US09106910B2

公开(公告)日：2015-08-11

申请号：US13380632

申请日：2010-06-21

Applicant: Stéphane Pateux ,  Nathalie Cammas ,  Isabelle Amonou

Inventor： Stéphane Pateux ,  Nathalie Cammas ,  Isabelle Amonou

IPC: H04N7/32 ,  H04N19/109 ,  H04N19/52 ,  H04N19/176 ,  H04N19/147 ,  H04N19/46 ,  H04N19/61

CPC classification number: H04N19/109 ,  H04N19/147 ,  H04N19/176 ,  H04N19/46 ,  H04N19/52 ,  H04N19/61

Abstract: A method and apparatus are provided for coding a signal from images, the images being split into sub-blocks of pixels. The method includes grouping at least two sub-blocks into at least one block of larger size, when the sub-blocks comply with at least one predetermined grouping criterion. A prediction is performed by applying at least one mode of motion prediction using at least two distinct motion prediction vectors, for at least one block of larger size, the motion prediction vectors being associated respectively with sub-sets of the block of larger size, comprising at least one of the sub-blocks of the block of larger size. The sub-sets are predefined and distinct.

Abstract translation: 提供了一种用于对来自图像的信号进行编码的方法和装置，所述图像被分割成像素的子块。 该方法包括当子块符合至少一个预定分组标准时，将至少两个子块分组成更大尺寸的至少一个块。 通过使用至少两个不同的运动预测向量对至少一个较大尺寸的块应用至少一种运动预测模式来执行预测，运动预测向量分别与较大尺寸的块的子集相关联，包括 块大小的子块中的至少一个。 子集是预定义和不同的。

公开(公告)号：US08743950B2

公开(公告)日：2014-06-03

申请号：US11996196

申请日：2006-07-18

Applicant: Isabelle Amonou ,  Stephane Pateux ,  Gerard Babonneau

Inventor： Isabelle Amonou ,  Stephane Pateux ,  Gerard Babonneau

IPC: H04N7/12 ,  H04N11/02 ,  H04N11/04 ,  G03B13/00 ,  G06F15/16

CPC classification number: H04N21/234381 ,  H04N21/234327 ,  H04N21/234354 ,  H04N21/234363 ,  H04N21/2662 ,  H04N21/4621 ,  H04N21/64792

Abstract: A method is provided for filtering a scalable video stream, organized in the form of data unit blocks, each of which comprises a base data unit and a set of data units distributed according to two types of enhancement data, corresponding respectively to time and/or space and/or quality characteristics and making it possible to define several quality levels. The method defines at least two distinct filtering profiles, or paths of each block data units, wherein each path defines a row of successive foldover positions and each foldover position uses at least by one less data units than a previous position, and selects one of the paths according to a predetermined criterion, taking into account the type of content of the stream and/or at least one information item representing the capacities of a terminal receiving the stream.

Abstract translation: 提供了一种用于过滤以数据单元块的形式组织的可伸缩视频流的方法，每个数据单元块包括分别对应于时间和/或时间的两种类型的增强数据的基本数据单元和一组数据单元， 空间和/或质量特征，并且可以定义几个质量水平。 该方法定义了至少两个不同的过滤简档或每个块数据单元的路径，其中每个路径定义一连续的后置位置行，并且每个舍弃位置使用比先前位置少至少一个数据单元，并且选择 考虑到流的内容的类型和/或表示接收流的终端的容量的至少一个信息项的根据预定标准的路径。

公开(公告)号：US08483493B2

公开(公告)日：2013-07-09

申请号：US12809838

申请日：2008-12-19

Applicant: Isabelle Amonou ,  Nathalie Cammas ,  Sylvain Kervadec ,  Stephane Pateux

Inventor： Isabelle Amonou ,  Nathalie Cammas ,  Sylvain Kervadec ,  Stephane Pateux

IPC: G06K9/36 ,  G06K9/46 ,  H04B1/66 ,  H04N7/12 ,  H04N11/02 ,  H04N11/04

CPC classification number: H04N19/523 ,  H04N19/103 ,  H04N19/117 ,  H04N19/132 ,  H04N19/156 ,  H04N19/18 ,  H04N19/46 ,  H04N19/61

Abstract: A method is provided decoding a digital image signal in a decoding terminal that uses a variable-complexity decoding, enabling, for at least one step in the decoding, selection of at least one decoding tool from at least two available decoding tools. The method includes the following steps: identifying from the images control images and intermediate images; applying to each of the control images, for at least one step of the decoding, at least one predetermined decoding tool imposed by the signal; applying to at least one of the intermediate images, for at least one step of the decoding, at least one decoding tool selected by the decoding terminal and not imposed by the signal.

Abstract translation: 提供一种在使用可变复杂度解码的解码终端中对数字图像信号进行解码的方法，使得至少一个步骤能够从至少两个可用的解码工具中选择至少一个解码工具。 该方法包括以下步骤：从图像中识别控制图像和中间图像; 对每个控制图像应用至少一个解码步骤，由该信号施加的至少一个预定解码工具; 对至少一个中间图像应用至少一个解码步骤，由解码终端选择并且不由该信号施加的至少一个解码工具。

公开(公告)号：US08315315B2

公开(公告)日：2012-11-20

申请号：US11813836

申请日：2005-12-20

Applicant: Isabelle Amonou ,  Nathalie Cammas ,  Stephane Pateux

Inventor： Isabelle Amonou ,  Nathalie Cammas ,  Stephane Pateux

IPC: H04N7/12

CPC classification number: H04N19/615 ,  H04N19/115 ,  H04N19/13 ,  H04N19/146 ,  H04N19/147 ,  H04N19/34 ,  H04N19/37 ,  H04N19/61 ,  H04N19/63

Abstract: A method is provided for encoding a sequence of images generating a data stream in the form of a structure of embedded data layers of n successive levels each of which corresponds to the predetermined resolution of said images. Said method comprises an encoding stage, which encodes at least one of said layers of an n+1 level by prediction on the basis of said level n layer and encodes each data layer of the level n in the form of a base sub-stream and, optionally, in the form of at least one enhancement sub-stream enabling to obtain at least one version of the enhanced quality of said images. The method also encodes, for at least one data layer of the n level, at least one single sub-stream enabling to reconstruct the version of said images at said level n resolution with a quality higher than said enhanced qualities, wherein said single sub-steam of the level n layer is not used during the prediction encoding of the layers of a level other than the level n.

Abstract translation: 提供了一种用于编码以n个连续级别的每个相应于所述图像的预定分辨率的嵌入数据层的结构的形式生成数据流的方法。 所述方法包括编码级，其通过基于所述级别n层的预测编码n + 1级的所述层中的至少一个层，并且以基本子流的形式对级别n的每个数据层进行编码， ，可选地，以至少一个增强子流的形式，使得能够获得所述图像的增强质量的至少一个版本。 该方法还对于n级的至少一个数据层编码至少一个单个子流，其能够以高于所述增强质量的质量以所述级别n分辨率重建所述图像的版本，其中所述单个子流 在级别n以外的层次的预测编码期间，不使用层次n层的蒸汽。

公开(公告)号：US20120263232A1

公开(公告)日：2012-10-18

申请号：US13500583

申请日：2010-10-01

Applicant: Stephane Pateux ,  Isabelle Amonou ,  Nathalie Cammas

Inventor： Stephane Pateux ,  Isabelle Amonou ,  Nathalie Cammas

IPC: H04N7/32

CPC classification number: H03M7/4006 ,  H04N19/196 ,  H04N19/197 ,  H04N19/436 ,  H04N19/91

Abstract: A method is provided for encoding an image represented by a sequence of symbols. The method includes the following steps: defining subsequences within the sequence; and going through the subsequences applying the following sub-steps for a routine subsequence: determining an adjacent routine encoded group including two previously encoded subsequences, one of which belongs to an adjacent encoded group associated with a subsequence that is separate from the routine subsequence; initializing a routine probability of occurrence of a symbol, taking into account two known probabilities of occurrence of the symbol, relating to two previously encoded subsequences of the adjacent routine encoded group; the consecutive entropic encoding of the symbols, and updating the routine probability of occurrence; and storing the routine probability of occurrence after encoding the last symbol.

Abstract translation: 提供了一种用于对由符号序列表示的图像进行编码的方法。 该方法包括以下步骤：定义序列内的子序列; 并且通过对于例程子序列应用以下子步骤的子序列：确定包括两个先前编码的子序列的相邻例程编码组，其中之一属于与子程序分离的子序列相关联的相邻编码组; 考虑到与相邻的例行编码组的两个先前编码的子序列有关的符号出现的两个已知概率，初始化符号发生的常规概率; 符号的连续熵编码，并更新常规发生概率; 并且在编码最后一个符号之后存储发生的常规概率。

公开(公告)号：US20120219066A1

公开(公告)日：2012-08-30

申请号：US13503553

申请日：2010-10-14

Applicant: Isabelle Amonou ,  Nathalie Cammas ,  Stephane Pateux ,  Stephanie Relier

Inventor： Isabelle Amonou ,  Nathalie Cammas ,  Stephane Pateux ,  Stephanie Relier

IPC: H04N7/26

CPC classification number: H04N19/436 ,  H04N19/174 ,  H04N19/70

Abstract: A method and apparatus are provided for encapsulating a plurality of data substreams encoded in a bit stream, wherein the encoded data substream is obtained by entropy encoding of a symbol subsequence representing an image or a series of images. The method includes grouping data substreams into substream pairs according to an iterative traversal order predetermined from the plurality of substreams, and on the basis of a pair of the substreams, obtaining a symmetrical pair by reversing a data order making up the second substream of the pair. The obtained pairs are then concatenated within a sequence. The binary stream is created by encapsulating the sequence and information related to a number of substreams and related to the sizes of pairs of symmetrical substreams.

Abstract translation: 提供了一种用于封装编码在比特流中的多个数据子流的方法和装置，其中通过对表示图像或一系列图像的符号子序列进行熵编码来获得编码数据子流。 该方法包括根据从多个子流预定的迭代遍历顺序将数据子流分组成子流对，并且基于一对子流，通过反转构成该对的第二子流的数据顺序来获得对称对 。 所获得的对然后在序列内连接。 通过封装与多个子流相关的序列和信息并与对称子流对的大小相关联来创建二进制流。

公开(公告)号：US20120183070A1

公开(公告)日：2012-07-19

申请号：US13380632

申请日：2010-06-21

Applicant: Stéphane Pateux ,  Nathalie Cammas ,  Isabelle Amonou

Inventor： Stéphane Pateux ,  Nathalie Cammas ,  Isabelle Amonou

IPC: H04N7/32

CPC classification number: H04N19/109 ,  H04N19/147 ,  H04N19/176 ,  H04N19/46 ,  H04N19/52 ,  H04N19/61

Abstract: A method and apparatus are provided for coding a signal from images, the images being split into sub-blocks of pixels. The method includes grouping at least two sub-blocks into at least one block of larger size, when the sub-blocks comply with at least one predetermined grouping criterion. A prediction is performed by applying at least one mode of motion prediction using at least two distinct motion prediction vectors, for at least one block of larger size, the motion prediction vectors being associated respectively with sub-sets of the block of larger size, comprising at least one of the sub-blocks of the block of larger size. The sub-sets are predefined and distinct.

Abstract translation: 提供了一种用于对来自图像的信号进行编码的方法和装置，所述图像被分割成像素的子块。 该方法包括当子块符合至少一个预定分组标准时，将至少两个子块分组成更大尺寸的至少一个块。 通过使用至少两个不同的运动预测向量对至少一个较大尺寸的块应用至少一种运动预测模式来执行预测，运动预测向量分别与较大尺寸的块的子集相关联，包括 块大小的子块中的至少一个。 子集是预定义和不同的。

公开(公告)号：US06795577B2

公开(公告)日：2004-09-21

申请号：US09726569

申请日：2000-12-01

Applicant: Isabelle Amonou

Inventor： Isabelle Amonou

IPC: G06K934

CPC classification number: G06T7/11 ,  G06T2207/10016 ,  G06T2207/20016 ,  G06T2207/20064

Abstract: The invention concerns a method of analyzing a set of data representing physical quantities, including the steps of: decomposition (E20) of the set of data on a plurality of resolution levels, first segmentation (E21) of at least one sub-part of a given resolution level, into at least two homogeneous regions, said given resolution level not being the highest resolution level in the decomposition, characterized in that it includes the steps of: extraction (E81) of contour data from the result of the segmentation of the previous steps, second segmentation (E25) of at least one sub-part of the resolution level higher than the given level into at least one homogeneous region, as a function of the contour data extracted.

Abstract translation: 本发明涉及一种分析表示物理量的一组数据的方法，包括以下步骤：在多个分辨率级别上对该组数据进行分解（E20），第一分割（E21）至少一个子部分 给定分辨率级别，进入至少两个均匀区域，所述给定分辨率级别不是分解中的最高分辨率级别，其特征在于，其包括以下步骤：从先前的分割结果的分割结果中提取（E81）轮廓数据 作为提取的轮廓数据的函数的步骤，将分辨率水平高于给定水平的至少一个子部分的第二分割（E25）转换为至少一个均匀区域。

公开(公告)号：US09055294B2

公开(公告)日：2015-06-09

申请号：US13146875

申请日：2010-01-26

Applicant: Isabelle Amonou ,  Matthieu Moinard ,  Nathalie Cammas

Inventor： Isabelle Amonou ,  Matthieu Moinard ,  Nathalie Cammas

IPC: H04N7/26 ,  H04N19/192 ,  H04N19/176 ,  H04N19/61 ,  H04N19/593 ,  H04N19/11 ,  H04N19/14 ,  H04N19/182

CPC classification number: H04N19/192 ,  H04N19/11 ,  H04N19/14 ,  H04N19/176 ,  H04N19/182 ,  H04N19/593 ,  H04N19/61

Abstract: A method is provided for encoding at least one image, divided into macroblocks, one macroblock including a set of blocks de pixels. The method includes, for at least one current macroblock of a current image, at least two iterations of the following steps: allocation of a priority level for encoding to at least one pixel of the current macroblock, selection of a pixel with the highest priority level called the priority pixel, prediction of a set of pixels including the priority pixel and a step of encoding a remainder from prediction for a region made up of predicted pixels of the macroblock, called the predicted region, the encoding step being used as and when the predicted region meets a predetermined encoding criterion.

Abstract translation: 提供了一种用于将至少一个图像分割成宏块的方法，一个宏块包括一组块de像素。 该方法包括对于当前图像的至少一个当前宏块，进行以下步骤的至少两次迭代：将编码优先级分配给当前宏块的至少一个像素，选择具有最高优先级的像素 称为优先像素，包括优先像素的一组像素的预测，以及将由预测的宏块的预测像素组成的区域预测的余数进行编码的步骤，称为预测区域，编码步骤用作何时以及何时 预测区域满足预定的编码标准。

公开(公告)号：US08971648B2

公开(公告)日：2015-03-03

申请号：US14291929

申请日：2014-05-30

Applicant: Isabelle Amonou ,  Stephane Pateux ,  Nathalie Cammas ,  Sylvain Kervadec

Inventor： Isabelle Amonou ,  Stephane Pateux ,  Nathalie Cammas ,  Sylvain Kervadec

IPC: G06K9/36 ,  G06K9/46

CPC classification number: H04N19/126 ,  H04N19/103 ,  H04N19/139 ,  H04N19/14 ,  H04N19/147 ,  H04N19/17 ,  H04N19/176 ,  H04N19/177 ,  H04N19/182 ,  H04N19/196 ,  H04N19/20 ,  H04N19/37 ,  H04N19/46 ,  H04N19/463 ,  H04N19/503 ,  H04N19/527 ,  H04N19/61 ,  H04N19/70

Abstract: A method and apparatus are provided for coding an image or a sequence of images, generating a data stream including data representative of pixel groups, referred to as blocks, in one of the images. The method includes: grouping blocks in a cluster of blocks according to the proximity of their respective values corresponding to at least one block parameter to be coded; determining a value of the parameter, the value being characteristic of said group of blocks; coding blocks of the cluster, where the values of the blocks for the parameter are coded implicitly by inheritance of the characteristic value or are coded as refinements relative to the characteristic value, and coding a data structure associated with the cluster of blocks, the data structure including data associated with the characteristic value.

Abstract translation: 提供了一种用于对图像或一系列图像进行编码的方法和装置，在其中一个图像中生成包括表示像素组的数据的数据流，称为块。 该方法包括：根据其对应于要编码的至少一个块参数的相应值的接近度来对块簇中的块进行分组; 确定参数的值，所述值是所述块组的特征; 编码块的编码块，其中参数的块的值通过特征值的继承隐含地编码，或者被编码为相对于特征值的细化，以及编码与块簇相关联的数据结构，数据结构 包括与特征值相关联的数据。