公开(公告)号：KR1020080114482A

公开(公告)日：2008-12-31

申请号：KR1020080015011

申请日：2008-02-19

Applicant: 삼성전자주식회사 ,  경희대학교 산학협력단

Inventor： 정도영 ,  박태성 ,  오윤제 ,  서덕영 ,  이영렬 ,  박광훈 ,  김규헌 ,  박민우 ,  임성창

IPC: H04N19/85 ,  H04N19/567 ,  H04N13/00

CPC classification number: H04N19/105 ,  H04N19/159 ,  H04N19/196 ,  H04N19/197 ,  H04N19/46 ,  H04N19/577 ,  H04N19/597

Abstract: A method and an apparatus for obtaining a brightness compensation prediction value and a brightness compensation flag value in a brightness compensation process during multi view coding are provided to estimate the brightness compensation prediction value and the brightness compensation flag value more effectively, thereby improving coding efficiency. When a current block performing brightness compensation is in a skip mode, a brightness compensation prediction value is obtained by using information of blocks adjacent to the current block(1010). When the obtained brightness compensation prediction value is 0, it is set that the brightness compensation of the current block is not performed(1030). When the brightness compensation prediction value is not 0, it is set that the brightness compensation of the current block is performed(1040).

Abstract translation: 提供一种用于在多视图编码期间的亮度补偿处理中获得亮度补偿预测值和亮度补偿标志值的方法和装置，以更有效地估计亮度补偿预测值和亮度补偿标志值，从而提高编码效率。 当执行亮度补偿的当前块处于跳过模式时，通过使用与当前块相邻的块的信息（1010）来获得亮度补偿预测值。 当获得的亮度补偿预测值为0时，设定当前块的亮度补偿不被执行（1030）。 当亮度补偿预测值不为0时，设定当前块的亮度补偿（1040）。

公开(公告)号：KR1020080114481A

公开(公告)日：2008-12-31

申请号：KR1020080014999

申请日：2008-02-19

Applicant: 삼성전자주식회사 ,  경희대학교 산학협력단

Inventor： 정도영 ,  박태성 ,  오윤제 ,  서덕영 ,  이영렬 ,  박광훈 ,  김규헌 ,  박민우 ,  임성창

IPC: H04N19/85 ,  H04N19/567 ,  H04N13/00

CPC classification number: H04N19/196 ,  H04N19/105 ,  H04N19/159 ,  H04N19/176 ,  H04N19/197 ,  H04N19/198 ,  H04N19/46 ,  H04N19/597 ,  H04N19/61

Abstract: A method and an apparatus for inducing a brightness compensation prediction value and a brightness compensation flag value in a brightness compensation process during multi view video coding are provided to predict a flag prediction value which indicates whether brightness compensation is performed during multi view coding and the brightness compensation prediction value more effectively, thereby improving coding efficiency. A brightness compensation prediction value is induced by using information of blocks adjacent to a current block when the current block for performing brightness compensation is in a B skip mode(510). When the induced brightness compensation prediction value is 0, it is set not to perform brightness compensation of the current block(530). When the induced brightness compensation prediction value is not 0, it is set to perform the brightness compensation of the current block(540).

Abstract translation: 提供了一种用于在多视图视频编码期间在亮度补偿处理中引发亮度补偿预测值和亮度补偿标志值的方法和装置，用于预测指示在多视图编码期间是否执行亮度补偿的标志预测值和亮度补偿 补偿预测值更有效，从而提高编码效率。 当用于执行亮度补偿的当前块处于B跳过模式时，通过使用与当前块相邻的块的信息来感生亮度补偿预测值（510）。 当感应亮度补偿预测值为0时，不设置当前块的亮度补偿（530）。 当感应亮度补偿预测值不为0时，设定为执行当前块的亮度补偿（540）。

公开(公告)号：KR1020080005848A

公开(公告)日：2008-01-15

申请号：KR1020070055942

申请日：2007-06-08

Applicant: 경희대학교 산학협력단

Inventor： 서덕영 ,  오재열 ,  박광훈 ,  박민우

IPC: H04N19/174 ,  H04N19/42

CPC classification number: H04N19/174 ,  H03K19/21 ,  H04N19/177 ,  H04N19/188 ,  H04N19/30

Abstract: A method for generating an XR(Xor Redundant) picture slice and a loss slice recovering method using the same are provided to recover a loss slice without an additional retransmission process although a slice is lost by generating an XR slice for important slices to be protected after compressing a moving picture and transmitting the generated XR slice. An XR slice application layer and an application range are set up(501). In the set XR slice application layer, the XOR operation of frames including at least one slice is performed in the set application range, and an XR slice is generated(502). When the generated XR slice is transmitted to a decoder, information of the set XR slice application layer and information of the set range are transmitted as a certain type.

Abstract translation: 提供了一种用于生成XR（Xor冗余）图像片和使用其的丢失片恢复方法的方法，以便在没有附加的重传过程的情况下恢复丢失片，尽管通过为重要片段生成XR片而丢失片， 压缩运动图像并发送生成的XR切片。 设置XR切片应用层和应用范围（501）。 在设置的XR片应用层中，在设定的应用范围内执行包含至少一个片的帧的XOR运算，生成XR片（502）。 当将生成的XR片段发送到解码器时，设置的XR片段应用层的信息和设置范围的信息作为某种类型被发送。

公开(公告)号：KR100794855B1

公开(公告)日：2008-01-14

申请号：KR1020060043338

申请日：2006-05-15

Applicant: 경희대학교 산학협력단

Inventor： 박광훈 ,  박민우 ,  서덕영

IPC: H04N19/597 ,  H04N13/00

Abstract: 본 발명은 다시점 비디오 코딩에 있어서 뷰 사이의 이미지 정보를 보상하는 과정에서 이미지 정보 변이값을 예측하는 방법과 이를 이용한 다시점 영상의 부호화/복호화 방법 및 장치에 관한 것이다. 본 발명의 바람직한 실시예에 따른 이미지 정보 변이값의 예측 방법에 의하면, 이미지 정보 변이값에 대한 예측 효율을 높이기 위하여 현재 블록과 동일한 영상 내에서 현재 블록에 인접한 코딩 완료된 블록 뿐만 아니라 같은 뷰 내에서 현재 영상과 다른 시간에서의 코딩 완료된 영상에서 현재 블록에 대응하는 위치의 블록 및 그 주변 블록의 이미지 정보 변이값을 이용하여 현재 블록에 대한 이미지 정보 변이값을 예측한다.
MPEG, H.264/AVC, 다시점 영상, 변이값 보상, 휘도 변이, 색차 변이, RGB 변이

公开(公告)号：KR1020080004392A

公开(公告)日：2008-01-09

申请号：KR1020070067031

申请日：2007-07-04

Applicant: 한국전자통신연구원 ,  경희대학교 산학협력단

Inventor： 정세윤 ,  박광훈 ,  박민우 ,  신승표 ,  서덕영 ,  문경애 ,  홍진우

IPC: H04N19/30 ,  H04N19/105 ,  H04N19/51

Abstract: A scalable video encoding/decoding method and an apparatus thereof are provided to override an existing weight given in a slice unit to a corresponding block of an enhancement layer of a reference frame with a greater weight than the existing weight in forming the reference block of the enhancement layer when the video data of a macro block of a basic layer is in a skip mode, thereby improving efficiency of SVC(Scalable Video Coding). A scalable video encoding method comprises the following steps of: determining whether the block of a basic layer corresponding to the block of an enhancement layer of a current frame to encode is in a skip mode(S210); overriding an existing weight set in the block of the enhancement layer of a reference frame corresponding to the block of the enhancement layer of the current frame with a new weight when the block of the basic layer is in the skip mode(S220); and generating a reference block for the block of the enhancement layer of the current frame based on the block of the enhancement layer of the reference frame and the block of the basic layer of the current frame by using the new weight(S230).

Abstract translation: 提供了一种可分级的视频编码/解码方法及其装置，以将在片单元中给出的现有权重覆盖参考帧的增强层的对应块，其中权重大于现有权重，以形成参考块的参考块 当基本层的宏块的视频数据处于跳过模式时，增强层，从而提高SVC（可缩放视频编码）的效率。 可扩展视频编码方法包括以下步骤：确定与要编码的当前帧的增强层的块相对应的基本层的块是否处于跳过模式（S210）; 当基本层的块处于跳过模式时，用当前帧的当前帧的增强层的块对应的参考帧的增强层的块中的现有权重集合（S220）; 并且通过使用新的权重，基于参考帧的增强层的块和当前帧的基本层的块来生成当前帧的增强层的块的参考块（S230）。

公开(公告)号：KR1020080004340A

公开(公告)日：2008-01-09

申请号：KR1020070040969

申请日：2007-04-26

Applicant: 한국전자통신연구원 ,  경희대학교 산학협력단

Inventor： 정세윤 ,  박광훈 ,  박민우 ,  신승표 ,  서덕영 ,  문경애 ,  홍진우

IPC: H04N19/30 ,  H04N19/103

CPC classification number: H04N19/34 ,  H04N19/132 ,  H04N19/157 ,  H04N19/176 ,  H04N19/196 ,  H04N19/61

Abstract: A method for scalable encoding of video data and an apparatus thereof are provided to override an existing weight value given in a slice unit with a greater weight when the reference block of an enhancement layer is generated if the video data of a macro block of a base layer is in a skip mode, thereby improving efficiency of the scalable video encoding. A method for scalable encoding of video data comprises the following steps of: generating a reference block by using a weight set for a corresponding block when the video data of a macro block can be predicted from the video data of reference pictures for the video data of the macro block(220); and generating the reference block by using a weight given to a slice belonging to a current frame when the video data of the macro block can not be predicted from the video data of the reference pictures.

Abstract translation: 提供一种用于视频数据的可分级编码的方法及其装置，用于当生成增强层的参考块时，以更大的权重来覆盖在片单元中给出的现有权重值，如果基址的宏块的视频数据 层处于跳过模式，从而提高可扩展视频编码的效率。 一种用于视频数据的可分级编码的方法包括以下步骤：当可从根据视频数据的视频数据的参考图像的视频数据预测宏块的视频数据时，通过使用对于相应块的权重来生成参考块 宏块（220）; 以及当所述宏块的视频数据不能从所述参考图像的视频数据中预测时，通过使用给予属于当前帧的切片的权重来生成所述参考块。

公开(公告)号：KR1020070117432A

公开(公告)日：2007-12-12

申请号：KR1020060132575

申请日：2006-12-22

Applicant: 경희대학교 산학협력단

Inventor： 박광훈 ,  박민우 ,  서덕영

IPC: H04N19/597 ,  H04N13/00

CPC classification number: H04N19/597 ,  H04N19/103 ,  H04N19/46 ,  H04N19/513 ,  H04N19/573 ,  H04N19/61 ,  H04N19/85

Abstract: A motion vector prediction method in multi-view video coding and an apparatus and a method for encoding/decoding a multi-view image using the same are provided to find a motion vector prediction value in a multi-view video coding so that the amount of information(entropy) is reduced, consequently the efficiency of prediction coding efficiency is improved. A motion vector prediction value is calculated by using at least one adjustably selected among motion vectors of neighboring blocks of a present block, motion vectors of blocks of corresponding to a time direction of the present block, and motion vectors of blocks corresponding to a view direction of the present block. The motion vector prediction value is the mean value of the selected motion vector. One motion vector is selected, and the motion vector prediction value is identical to the selected motion vector. The motion vector prediction value is the mean value of the selected motion vector and the vector(0,0).

Abstract translation: 提供多视点视频编码中的运动矢量预测方法以及使用其的多视点图像的编码/解码方法，以便在多视点视频编码中找到运动矢量预测值， 信息（熵）减少，因此提高了预测编码效率。 通过使用当前块的相邻块的运动矢量中的至少一个可选择地选择运动矢量预测值，对应于当前块的时间方向的块的运动矢量和对应于视图方向的块的运动矢量 的当前块。 运动矢量预测值是所选运动矢量的平均值。 选择一个运动矢量，运动矢量预测值与选择的运动矢量相同。 运动矢量预测值是所选运动矢量和向量（0,0）的平均值。

公开(公告)号：KR1020070117424A

公开(公告)日：2007-12-12

申请号：KR1020060087176

申请日：2006-09-11

Applicant: 경희대학교 산학협력단

Inventor： 서덕영 ,  김철근 ,  박광훈

IPC: H04N19/597 ,  H04N13/00

CPC classification number: H04N19/597 ,  G06T5/40 ,  G06T7/97 ,  H04N19/159

Abstract: A method for compensating for image information disparity through contrast correction and a method and an apparatus for decoding/encoding a multi-view video by using the compensating method are provided to improve a prediction coding efficiency when an average difference or a contrast difference exists in the image information of two images. The contrast of a reference image having a view different from that of a present image is corrected(S21). A prediction image of the present image is created by using image information of the corrected reference image. The contrast is corrected by using contrast correction information of the input image. The contrast correction information of the image includes information about a correction scheme. The image information of the prediction image is identical to the image information of the reference image.

Abstract translation: 提供了一种通过对比度校正来补偿图像信息差异的方法，以及通过使用补偿方法对多视点视频进行解码/编码的方法和装置，以提供在当前平均差异或对比度差异存在时的预测编码效率 两幅图像的图像信息。 校正具有不同于当前图像的视图的参考图像的对比度（S21）。 通过使用校正的参考图像的图像信息来创建当前图像的预测图像。 通过使用输入图像的对比度校正信息来校正对比度。 图像的对比度校正信息包括关于校正方案的信息。 预测图像的图像信息与参考图像的图像信息相同。

公开(公告)号：KR100775787B1

公开(公告)日：2007-11-13

申请号：KR1020060071731

申请日：2006-07-28

Applicant: 경희대학교 산학협력단

Inventor： 서덕영 ,  박광훈

IPC: H04N19/137 ,  H04N19/177 ,  H04N19/176 ,  H04N19/30

Abstract: 1. 청구범위에 기재된 발명이 속한 기술분야
본 발명은, 영역별 시공간적 특성을 이용한 동영상 부호화 장치 및 그 방법에 관한 것임.
2. 발명이 해결하려고 하는 기술적 과제
본 발명은, GOP(group of pictures) 단위의 영상을 공간적 계층의 수와 시간적 해상도의 수에 따라 소정 개수의 매크로블록을 가지는 슈퍼 매크로블록 단위로 구분하고, 시공간적 활성도를 산출하여 소정 범위를 만족시키는 슈퍼 매크로블록별로 그룹핑한 후 해당 영역에 상응하는 부호화 방식으로 부호화함으로써, 압축의 효율성을 향상시키기 위한 영역별 시공간적 특성을 이용한 동영상 부호화 장치 및 그 방법을 제공하는데 그 목적이 있음.
3. 발명의 해결방법의 요지
본 발명은, 영역별 시공간적 특성을 이용한 동영상 부호화 장치에 있어서, GOP(group of pictures) 단위의 영상을 입력받기 위한 영상 입력수단; 상기 영상 입력수단에서 입력받은 영상을 공간적 계층의 수와 시간적 해상도의 수에 따라 소정 개수의 매크로블록을 가지는 슈퍼 매크로블록 단위로 구분하기 위한 슈퍼 매크로블록 구분수단; 상기 슈퍼 매크로블록 구분수단에서 구분한 슈퍼 매크로블록의 시공간적 활성도를 산출하기 위한 활성도 산출수단; 상기 활성도 산출수단에서 산출한 시공간적 활성도가 소정 범위를 만족시키는 슈퍼 매크로블록별로 그룹핑하기 위한 그룹핑수단; 및 상기 그룹핑수단에서 그룹핑한 슈퍼 매크로블록을 시공간적 활성도에 따라 부호화하기 위한 부호화수단을 포함함.
4. 발명의 중요한 용도
본 발명은 동영상의 부호화 등에 이용됨.
동영상 부호화, 슈퍼 매크로블록, 슈퍼 슬라이스, 시공간적 활성도, 비디오시퀀스

Abstract translation: 1.权利要求书中描述的发明所属的技术领域

公开(公告)号：KR100775422B1

公开(公告)日：2007-11-12

申请号：KR1020060066278

申请日：2006-07-14

Applicant: 경희대학교 산학협력단

Inventor： 서덕영

IPC: H04N7/24

CPC classification number: H04N19/42 ,  H04N19/122 ,  H04N19/132 ,  H04N19/176 ,  H04N19/46

Abstract: A method for compressing/decompressing an image irregularly sampled is provided to perform compression/decompression efficiently without an additional interpolation process by extracting a plurality of pixels from the image irregularly sampled, designating the order of the extracted pixels, mapping and compressing the pixels to an orthogonal vector, and decompressing the compressed pixels by using mapping type information and orthogonal vector set information. A plurality of pixels is extracted from an image irregularly sampled by using a square block(501). A plurality of order designation types is applied to the extracted pixels to designate the order of the respective pixels(502). An inner product of orthogonal vector values corresponding to the designated pixels values is performed for the pixels whose order is designated through the respective order designation types to calculate result values(503). After a result value with high energy concentration is searched among the result values corresponding to the respective order designation types, corresponding order designation type information and orthogonal vector information are inserted into transmission data(504). The pixels are decompressed by using the order designation type information and the orthogonal vector information(505).

Abstract translation: 提供用于压缩/解压缩不规则采样的图像的方法，以通过从不规则采样的图像中提取多个像素，指定提取的像素的顺序，将像素映射和压缩到一个像素来有效地执行压缩/解压缩 正交矢量，并使用映射类型信息和正交矢量集信息解压缩压缩像素。 通过使用方块（501）从不规则采样的图像中提取多个像素。 多个顺序指定类型被应用于提取的像素以指定各个像素的顺序（502）。 对于通过各个顺序指定类型指定顺序的像素执行与指定的像素值对应的正交矢量值的内积，以计算结果值（503）。 在对应于各个顺序指定类型的结果值中搜索具有高能量集中的结果值之后，将对应的顺序指定类型信息和正交向量信息插入到发送数据（504）中。 通过使用顺序指定类型信息和正交矢量信息来解压缩像素（505）。