公开(公告)号：US08149922B2

公开(公告)日：2012-04-03

申请号：US11665848

申请日：2005-10-21

申请人： Euee-S. Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong Ho Cho ,  Chung-Ku Lee

发明人： Euee-S. Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong Ho Cho ,  Chung-Ku Lee

IPC分类号： H04N7/12

CPC分类号： H04N19/12 ,  H04N19/44 ,  H04N19/61 ,  H04N19/70

摘要： A video unified codec device and corresponding video method. The unified codec device includes parsing and decoding functional units (PD FUs) extracting and grouping context information, control signals, and data in bit streams inputted according to different syntax data per codec, macro-block-based functional units (MB-based FUs) comprising block-based process units of each codec, for decoding data outputted from PD FUs, and a global control unit (GCU) for controlling MB-based FUs grouped after corresponding control signals and context information received from PD FUs to each codec and processing.

摘要翻译： 视频统一编解码器设备及相应的视频方式。 统一编解码器装置包括解码和解码功能单元（PD FU），提取和分组根据每个编解码器的不同语法数据输入的比特流的上下文信息，控制信号和数据，基于宏块的功能单元（基于MB的功能单元） 包括用于解码从PDFU输出的数据的每个编解码器的基于块的处理单元，以及用于控制在对应的控制信号之后分组的基于MB的FU以及从PDFU接收的上下文信息到每个编解码器和处理的全局控制单元（GCU）。

公开(公告)号：US20060109911A1

公开(公告)日：2006-05-25

申请号：US11256188

申请日：2005-10-24

申请人： Euee-S Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

发明人： Euee-S Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

IPC分类号： H04N11/04 ,  H04N7/12 ,  H04B1/66 ,  H04N11/02

CPC分类号： H04N19/18 ,  H04N19/11 ,  H04N19/12 ,  H04N19/15 ,  H04N19/176 ,  H04N19/46 ,  H04N19/593 ,  H04N19/61

摘要： There are provided selective prediction encoding and decoding methods and selective prediction encoding and decoding devices. The selective prediction encoding device selects and performs one of an AC/DC prediction encoding method and an intra prediction encoding method which corresponds to the smaller of code amount from AC/DC prediction and an amount of AVC intra coding, records information indicating the selected encoding method in a header of a bit stream, and transmits the bit stream to the selective prediction decoding device. The selective prediction decoding device decodes the transmitted bit stream by the use of a decoding process corresponding to the information recorded in the header. Accordingly, it is possible to improve compression ratio and image quality by using the selective prediction encoding and decoding methods.

摘要翻译： 提供选择性预测编码和解码方法以及选择性预测编码和解码装置。 选择性预测编码装置选择并执行AC / DC预测编码方式和对应于AC / DC预测的代码量越小和AVC帧内编码量越小的帧内预测编码方式之一，记录表示选择编码的信息 方法，并将比特流发送到选择性预测解码装置。 选择性预测解码装置通过使用与记录在头部中的信息相对应的解码处理对发送的比特流进行解码。 因此，可以通过使用选择性预测编码和解码方法来提高压缩率和图像质量。

公开(公告)号：US07912131B2

公开(公告)日：2011-03-22

申请号：US11256188

申请日：2005-10-24

申请人： Euee-S Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

发明人： Euee-S Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sung-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

IPC分类号： H04N11/02

CPC分类号： H04N19/18 ,  H04N19/11 ,  H04N19/12 ,  H04N19/15 ,  H04N19/176 ,  H04N19/46 ,  H04N19/593 ,  H04N19/61

摘要： There are provided selective prediction encoding and decoding methods and selective prediction encoding and decoding devices. The selective prediction encoding device selects and performs one of an AC/DC prediction encoding method and an intra prediction encoding method which corresponds to the smaller of code amount from AC/DC prediction and an amount of AVC intra coding, records information indicating the selected encoding method in a header of a bit stream, and transmits the bit stream to the selective prediction decoding device. The selective prediction decoding device decodes the transmitted bit stream by the use of a decoding process corresponding to the information recorded in the header. Accordingly, it is possible to improve compression ratio and image quality by using the selective prediction encoding and decoding methods.

摘要翻译： 提供选择性预测编码和解码方法以及选择性预测编码和解码装置。 选择性预测编码装置选择并执行AC / DC预测编码方式和对应于AC / DC预测的代码量越小和AVC帧内编码量越小的帧内预测编码方式之一，记录表示选择编码的信息 方法，并将比特流发送到选择性预测解码装置。 选择性预测解码装置通过使用与记录在头部中的信息相对应的解码处理对发送的比特流进行解码。 因此，可以通过使用选择性预测编码和解码方法来提高压缩率和图像质量。

公开(公告)号：US20080037649A1

公开(公告)日：2008-02-14

申请号：US11665848

申请日：2005-10-21

申请人： Euee-S. Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sun-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

发明人： Euee-S. Jang ,  Yung-Lyul Lee ,  Sun-Young Lee ,  Sun-Won Park ,  Jong-Woo Won ,  Yong-Ho Cho ,  Chung-Ku Lee

IPC分类号： H04N7/24 ,  H04N7/12

CPC分类号： H04N19/12 ,  H04N19/44 ,  H04N19/61 ,  H04N19/70

摘要： The present invention is directed to a video unified codec device and its method. According to an embodiment of this invention, the unified codec device comprises parsing and decoding functional units (PD FUs) extracting and grouping context information, control signals, and data in bit streams inputted according to different syntax data per codec, macro-block-based functional units (MB-based FUs) unified based on block-based process units of each codec, for decoding data outputted from PD FUs, and a global control unit (GCU) for controlling MB-based FUs grouped after corresponding control signals and context information received from PD FUs to each codec and processing. By this invention, a new concept and structure of unified codec corresponding to similarities, differences, and considerations between different codecs can be presented.

摘要翻译： 本发明涉及一种视频统一编解码器装置及其方法。 根据本发明的实施例，统一编解码器装置包括解码和解码功能单元（PD FU），提取和分组根据每个编解码器的不同语法数据输入的比特流的上下文信息，控制信号和数据，基于宏块 基于每个编解码器的基于块的处理单元统一的功能单元（基于MB的FU），用于解码从PD FU输出的数据;以及全局控制单元（GCU），用于控制在对应的控制信号和上下文信息之后分组的基于MB的FU 从PD FUs接收到每个编解码器和处理。 通过本发明，可以呈现对应于不同编解码器之间的相似性，差异和考虑的统一编解码器的新概念和结构。

公开(公告)号：US20060133501A1

公开(公告)日：2006-06-22

申请号：US11289650

申请日：2005-11-30

申请人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

IPC分类号： H04N11/02 ,  H04N11/04 ,  H04B1/66 ,  H04N7/12

CPC分类号： H04N19/51 ,  H04N19/61

摘要： Provided are a motion estimation device and method adaptive to change in illumination. The motion estimation method includes the steps of: generating a current frame pattern block for a current frame block and a reference frame pattern block for a reference frame block; calculating the sum of absolute differences (SAD) for candidate pattern blocks corresponding to the current frame pattern block among the reference frame pattern blocks and determining the candidate pattern block corresponding to the SAD satisfying a predetermined condition among the calculated SADs as a motion vector; encoding the current frame block, by generating a residual signal using a difference signal between the candidate pattern block corresponding to the motion vector and the current frame pattern block then performing discrete cosine transform and quantization to the residual signal with a predetermined encoding mode; and adding flag information indicating the addition of a mean pixel value applied to the encoded current frame block and identification information on the encoding mode as header information corresponding to the encoded current frame block. By omitting insertion of a mean pixel value for some fields, it is possible to minimize the amount of bit streams transmitted to a decoder.

摘要翻译： 提供了适应于照明变化的运动估计装置和方法。 运动估计方法包括以下步骤：产生用于当前帧块的当前帧模式块和用于参考帧块的参考帧模式块; 计算与参考帧图案块中的当前帧图案块相对应的候选图案块的绝对差（SAD）之和，并将所计算的SAD中满足预定条件的SAD对应的候选图案块确定为运动矢量; 通过使用与运动矢量相对应的候选模式块与当前帧模式块之间的差信号产生残差信号，然后以预定编码模式对剩余信号执行离散余弦变换和量化，对当前帧块进行编码; 以及添加指示将加到编码的当前帧块上的平均像素值的加法和编码模式的识别信息相加的标志信息作为与编码的当前帧块相对应的标题信息。 通过省略对某些场的平均像素值的插入，可以最小化发送到解码器的比特流的量。

公开(公告)号：US20080130738A1

公开(公告)日：2008-06-05

申请号：US11813969

申请日：2005-12-13

申请人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

IPC分类号： H04N11/02

CPC分类号： H04N19/436 ,  H04N19/105 ,  H04N19/136 ,  H04N19/172 ,  H04N19/177 ,  H04N19/597 ,  H04N19/61

摘要： A method for setting the order of reference frames for multi-view coding and a recorded medium recording the method are disclosed. The method of setting the order of reference frames, in accordance with an embodiment of the present invention, comprises the steps of selecting the n (integer) number of frames of the same kind that have a same-view relation with said current frame and are encoded or decoded prior to said current frame as a first reference frame, selecting the m (integer) number of frames of the same kind that have a same-view relation with said current frame and have an inter-view relation with a frame of the same kind, which is encoded or decoded immediately before, as a second reference frame, wherein the frame of the same kind has a same-view relation with said current frame, and encoding or decoding said current frame by making reference to the frames of the same kind included in said first reference frame and said second reference frame. With the present invention, an improved picture quality and three-dimensionality can be provided, and the delay in processing time can be prevented when using the reference frame.

摘要翻译： 公开了一种用于设置用于多视图编码的参考帧的顺序的方法和记录该方法的记录介质。 根据本发明的实施例，设置参考帧的顺序的方法包括以下步骤：选择与所述当前帧具有相同视图关系的相同种类的n（整数）个帧数，并且是 在所述当前帧之前被编码或解码为第一参考帧，选择与所述当前帧具有相同视图关系的相同种类的m（整数）个帧，并且与 相同种类，其被紧密地编码或解码为第二参考帧，其中相同类型的帧与所述当前帧具有相同视图关系，并且通过参考帧的帧来对所述当前帧进行编码或解码 包括在所述第一参考帧和所述第二参考帧中的相同种类。 利用本发明，可以提供改进的图像质量和三维度，并且当使用参考帧时可以防止处理时间的延迟。

公开(公告)号：US09042439B2

公开(公告)日：2015-05-26

申请号：US11813969

申请日：2005-12-13

申请人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

IPC分类号： H04N11/02 ,  H04N19/436 ,  H04N19/597 ,  H04N19/105 ,  H04N19/172 ,  H04N19/61 ,  H04N19/136 ,  H04N19/177

CPC分类号： H04N19/436 ,  H04N19/105 ,  H04N19/136 ,  H04N19/172 ,  H04N19/177 ,  H04N19/597 ,  H04N19/61

摘要： A method of setting an order of reference frames for multi-view coding performed by one or more processors. The method includes selecting one or more frames of the same kind that have a same-view relation with said current frame and have an inter-view relation with a frame of the same kind, which is encoded or decoded immediately before; and encoding or decoding the current frame by making reference to the selected one or more kind. Further, the one or more frames of the same kind is a P frame.

摘要翻译： 一种设置由一个或多个处理器执行的多视图编码的参考帧的顺序的方法。 该方法包括选择与所述当前帧具有相同视图关系的相同类型的一个或多个帧，并且具有与之前被编码或解码的相同类型的帧的视角间关系; 并且通过参考所选择的一种或多种类型对当前帧进行编码或解码。 此外，同一种类的一个或多个帧是P帧。

公开(公告)号：US08059720B2

公开(公告)日：2011-11-15

申请号：US11289651

申请日：2005-11-30

申请人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

IPC分类号： H04B7/12 ,  H04B1/66

CPC分类号： H04N19/59 ,  H04N19/107 ,  H04N19/124 ,  H04N19/147 ,  H04N19/176 ,  H04N19/40 ,  H04N19/51 ,  H04N19/513 ,  H04N19/61

摘要： In an image down-sampling transcoding method, by reducing an MPEG SP (simple profile) video bit stream having a CIF (Common Intermediate Format) size into a half in image size in a pixel domain, the video bit stream is converted into a H.264 BP (Baseline Profile) video stream having a QCIF (Quarter CIF) size. Accordingly, it is possible to down-sample an image from MPEG-4 to H.264 at a high speed.

摘要翻译： 在图像下采样代码转换方法中，通过在像素域中将具有CIF（公共中间格式）大小的MPEG SP（简单简档）视频比特流减小到图像大小的一半，将视频比特流转换为H .264具有QCIF（Quarter CIF）大小的BP（Baseline Profile）视频流。 因此，可以高速地将图像从MPEG-4下采样到H.264。

公开(公告)号：US07924923B2

公开(公告)日：2011-04-12

申请号：US11289650

申请日：2005-11-30

申请人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S Jang ,  Chung-Ku Lee

IPC分类号： H04N7/12 ,  H04N11/04

CPC分类号： H04N19/51 ,  H04N19/61

摘要： Provided are a motion estimation device and method adaptive to change in illumination. The motion estimation method includes the steps of: generating a current frame pattern block and a reference frame pattern block for a reference frame block; calculating the sum of absolute differences (SAD) for candidate pattern blocks and determining the candidate pattern block corresponding to the SAD satisfying a predetermined condition among the calculated SAD as a motion vector; encoding the current frame block, and adding flag information indicating the addition of a mean pixel value applied to the encoded current frame block and identification information on the encoding mode as header information corresponding to the encoded current frame block.

摘要翻译： 提供了适应于照明变化的运动估计装置和方法。 运动估计方法包括以下步骤：产生用于参考帧块的当前帧图案块和参考帧图案块; 计算候选模式块的绝对差值（SAD），并将所计算的SAD中满足预定条件的SAD对应的候选模式块确定为运动矢量; 编码当前帧块，以及添加指示加到编码的当前帧块上的平均像素值的相加标志信息和编码模式的识别信息作为与编码的当前帧块相对应的标题信息。

公开(公告)号：US20060133500A1

公开(公告)日：2006-06-22

申请号：US11289649

申请日：2005-11-30

申请人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

发明人： Yung-Lyul Lee ,  Euee-S. Jang ,  Chung-Ku Lee

IPC分类号： H04N11/02 ,  H04N7/12 ,  H04N11/04 ,  H04B1/66

CPC分类号： H04N19/59 ,  H04N19/103 ,  H04N19/157 ,  H04N19/159 ,  H04N19/176 ,  H04N19/196 ,  H04N19/40 ,  H04N19/513 ,  H04N19/56

摘要： Provided are method and device for down-sampling an image from H.264 to MPEG-4 at a high speed. The image down-sampling transcoding method for converting a first macro block having a first format into a second macro block having a second format includes: (a) determining a block mode of the second macro block and converting the first macro block into the second macro block with the determined block mode, when the number of intra macro blocks included in the first macro block satisfies a first condition; (b) determining a block mode of the second macro block and converting the first macro block into the second macro block with the determined block mode, when the number for inter macro blocks included in the first macro block satisfies a second condition; and (c) re-adjusting a motion vector of the second macro block, after the conversion of the first macro block into the second macro block is finished.

摘要翻译： 提供了用于从H.264到MPEG-4高速下采样图像的方法和装置。 用于将具有第一格式的第一宏块转换成具有第二格式的第二宏块的图像下采样转码方法包括：（a）确定第二宏块的块模式并将第一宏块转换为第二宏块 当包含在第一宏块中的宏块内的数量满足第一条件时，确定块模式的块; （b）当包含在第一宏块中的宏块之间的数量满足第二条件时，确定第二宏块的块模式并将第一宏块转换为具有所确定的块模式的第二宏块; 以及（c）在将第一宏块转换为第二宏块之后，重新调整第二宏块的运动矢量。