公开(公告)号：KR101365571B1

公开(公告)日：2014-02-21

申请号：KR1020070005807

申请日：2007-01-18

Applicant: 삼성전자주식회사

Inventor： 한우진 ,  이재출

IPC: H04N19/11 ,  H04N19/51

Abstract: 본 발명은 영상의 인트라 예측 부호화, 복호화 방법 및 장치에 관한 것으로, 본 발명에 따른 영상의 인트라 예측 부호화 방법은 현재 블록으로부터 x축 방향으로 N 픽셀, y축 방향으로 M 픽셀만큼 이동한 위치에 있는 참조 블록의 인트라 예측 방향에 기초하여 현재 블록의 인트라 예측 방향을 예측하고, 실제 현재 블록의 인트라 예측 방향 및 예측된 인트라 예측 방향에 기초하여 상기 현재 블록을 인트라 예측 부호화함으로써, 인트라 예측 방향을 보다 정확하게 예측할 수 있어 영상 데이터의 압축률이 향상된다.
인트라 예측, 방향, 참조 블록

公开(公告)号：KR1020130119396A

公开(公告)日：2013-10-31

申请号：KR1020130108052

申请日：2013-09-09

Applicant: 삼성전자주식회사

Inventor： 이태미 ,  한우진

IPC: H04N19/50

CPC classification number: H04N19/521 ,  H04N19/105 ,  H04N19/137 ,  H04N19/14 ,  H04N19/159 ,  H04N19/176 ,  H04N19/182 ,  H04N19/51 ,  H04N19/513 ,  H04N19/52 ,  H04N19/593 ,  H04N19/61

Abstract: PURPOSE: A method for estimating a motion vector by using a plurality of motion vector predictors and an apparatus thereof, an encoder, a decoder and a decoding method are provided to transmit partition information for information about motion vector difference and restoring of a motion vector of a current block. CONSTITUTION: A motion estimating unit generates a motion vector of each of a plurality of partitions adjacent to a current block and the current block. A motion vector estimating unit (503) selects at least one partition compared with the current block in order to acquire motion vector difference among a plurality of partitions adjacent to the current block on the basis of spatial similarity between the current block and the plurality of partitions adjacent to the current block. The motion vector estimating unit generates the motion vector difference. The motion vector estimating unit transmits partition information to restore the motion vector of the current block and the information about the motion vector difference. [Reference numerals] (501) Space similarity calculating unit; (502) Partition(s) selecting unit; (503) Motion vector estimating unit; (AA) Present frame; (BB) Reference frame

Abstract translation: 目的：提供一种通过使用多个运动矢量预测器及其装置，编码器，解码器和解码方法来估计运动矢量的方法，用于传送关于运动矢量差的信息和关于运动矢量的运动矢量的恢复的分区信息 一个当前块。 构成：运动估计单元生成与当前块和当前块相邻的多个分区中的每一个的运动矢量。 运动矢量估计单元（503）与当前块相比选择至少一个分区，以便根据当前块和多个分区之间的空间相似度来获取与当前块相邻的多个分区之间的运动矢量差 与当前块相邻。 运动矢量估计单元产生运动矢量差。 运动矢量估计单元发送分区信息以恢复当前块的运动矢量和关于运动矢量差的信息。 （附图标记）（501）空间相似度计算单元; （502）分区选择单元; （503）运动矢量估计单元; （AA）现有框架; （BB）参考框架

公开(公告)号：KR1020130092526A

公开(公告)日：2013-08-20

申请号：KR1020130079912

申请日：2013-07-08

Applicant: 삼성전자주식회사

Inventor： 이교혁 ,  한우진 ,  이태미

IPC: H04N19/50

CPC classification number: H04N19/52 ,  H04N19/132 ,  H04N19/176

Abstract: PURPOSE: A coding and decoding method of images and a device thereof are provided to save the number of bit needed for coding of the information about a motion vector by accurately predicting the motion vector of a current block and reducing difference values between the motion vector and the predicted motion vector. CONSTITUTION: An encoding device searches a reference picture by using pixels of a current block and generates a motion vector of the current block (510). The encoding device generates a predicted motion vector by searching the reference picture with the previously encoded pixels which are adjacent to the current block (520). The encoding device encodes the current block based on the predicted motion vector and the generated motion vector (530). [Reference numerals] (510) Generate a motion vector by searching a reference picture with pixels of a current block; (520) Generate a predicted motion vector, which is a predicted value of a motion vector, by searching a reference picture with previously encoded pixels which are adjacent to a current block; (530) Encode a current block based on a predicted motion vector and a motion vector; (AA) Start; (BB) End

Abstract translation: 目的：提供图像的编码和解码方法及其装置，以通过精确地预测当前块的运动矢量来减少对运动矢量的信息的编码所需的位数，并减少运动矢量与 预测运动矢量。 构成：编码装置通过使用当前块的像素搜索参考图像，并生成当前块的运动矢量（510）。 编码装置通过使用与当前块相邻的先前编码的像素搜索参考图像来生成预测的运动矢量（520）。 编码装置基于预测的运动矢量和产生的运动矢量（530）对当前块进行编码。 （附图标记）（510）通过使用当前块的像素搜索参考图片来生成运动矢量; （520）通过使用与当前块相邻的先前编码的像素搜索参考图像来生成作为运动矢量的预测值的预测运动矢量; （530）基于预测的运动矢量和运动矢量对当前块进行编码; （AA）开始; （BB）结束

公开(公告)号：KR1020130029361A

公开(公告)日：2013-03-22

申请号：KR1020120102405

申请日：2012-09-14

Applicant: 삼성전자주식회사

Inventor： 알쉬나엘레나 ,  천지엔러 ,  한우진

IPC: H04N19/103 ,  H04N19/50

CPC classification number: H04N19/105 ,  H04N19/107 ,  H04N19/109 ,  H04N19/136 ,  H04N19/146 ,  H04N19/176 ,  H04N19/182 ,  H04N19/19 ,  H04N19/503 ,  H04N19/513 ,  H04N19/523 ,  H04N19/57 ,  H04N19/61

Abstract: PURPOSE: A video encoding and decoding method and a device thereof are provided to predict and compensate for motions by determining a single direction or a both directions motion prediction mode based on a size of a prediction unit. CONSTITUTION: A motion prediction unit predicts a motion motions according to a single direction or a dual directions prediction mode. A compensation unit compensates for the motions according to the single direction or the dual directions prediction mode(2620). A control unit determines the motion prediction mode of the current prediction unit based on the acquired encoding cost of the current prediction unit(2630). An entropy encoding unit encodes the determined motion prediction mode as entropy based on the size of the current prediction unit(2640). [Reference numerals] (2610) Determining the use of a single directional motion prediction mode and a dual directional motion prediction mode to be applied to the motion prediction and compensation of a current prediction unit to be encoded based on the size of the current prediction unit; (2620) Executing motion prediction and compensation to a current prediction unit by applying at least one motion prediction mode of a single directional motion prediction mode and a bidirectional motion prediction mode according to the use of the single directional motion prediction mode and the bidirectional motion prediction mode; (2630) Determining the final motion prediction mode of the current prediction unit based on the encoding cost of the current prediction unit acquired by the motion prediction and compensation; (2640) Encoding determined final motion prediction mode information; (AA) Start; (BB) End

Abstract translation: 目的：提供视频编码和解码方法及其装置，以通过基于预测单元的大小确定单方向或双向运动预测模式来预测和补偿运动。 构成：运动预测单元根据单一方向或双向预测模式来预测运动运动。 补偿单元根据单向或双向预测模式补偿运动（2620）。 控制单元基于所获取的当前预测单元（2630）的编码成本来确定当前预测单元的运动预测模式。 熵编码单元基于当前预测单元（2640）的大小将所确定的运动预测模式编码为熵。 （附图标记）（2610）基于当前预测单元的大小来确定要应用于要被编码的当前预测单元的运动预测和补偿的单向运动预测模式和双向运动预测模式的使用 ; （2620）根据单向运动预测模式和双向运动预测的使用，通过应用单向运动预测模式和双向运动预测模式的至少一个运动预测模式，对当前预测单元执行运动预测和补偿 模式; （2630）基于通过运动预测和补偿获取的当前预测单元的编码成本确定当前预测单元的最终运动预测模式; （2640）对确定的最终运动预测模式信息进行编码; （AA）开始; （BB）结束

公开(公告)号：KR1020130004181A

公开(公告)日：2013-01-09

申请号：KR1020120071974

申请日：2012-07-02

Applicant: 삼성전자주식회사

Inventor： 이태미 ,  천지엔러 ,  한우진

IPC: H04N19/50 ,  H04N19/103 ,  H04N19/70

CPC classification number: H04N19/91 ,  H04N19/103 ,  H04N19/105 ,  H04N19/46 ,  H04N19/52

Abstract: PURPOSE: A method and an apparatus for encoding and decoding motion information using a skip mode are provided to improve the coding efficiency of a skip mode. CONSTITUTION: A motion information encoding apparatus obtains multiple motion information candidates about a current prediction unit by using prediction units. The apparatus determines a most probable predictor candidate among multiple motion information candidates. The apparatus encodes a flag in a highest probable skip mode by using the most probable predictor candidate. [Reference numerals] (110) Motion information candidate group generation unit; (120) Best probability predicted candidate determining unit; (130) Motion information encoding unit; (AA) Bit stream

Abstract translation: 目的：提供一种使用跳过模式对运动信息进行编码和解码的方法和装置，以提高跳过模式的编码效率。 构成：运动信息编码装置通过使用预测单位获取关于当前预测单元的多个运动信息候选。 该装置在多个运动信息候选中确定最可能的预测器候选。 该装置通过使用最可能的预测器候选来编码具有最高可能跳过模式的标志。 （附图标记）（110）运动信息候补组生成部; （120）最佳概率预测候选决定单元; （130）运动信息编码单元; （AA）位流

公开(公告)号：KR1020120038365A

公开(公告)日：2012-04-23

申请号：KR1020110100036

申请日：2011-09-30

Applicant: 삼성전자주식회사

Inventor： 천민수 ,  민정혜 ,  한우진 ,  이태미 ,  이선일

IPC: H04N19/30 ,  H04N19/103

CPC classification number: H04N19/30 ,  H04N19/119 ,  H04N19/176 ,  H04N19/463 ,  H04N19/61 ,  H04N19/96 ,  H04N19/102 ,  H04N19/103 ,  H04N19/169 ,  H04N19/186 ,  H04N19/70

Abstract: PURPOSE: A video encoding/decoding method for encoding/decoding a symbol in a hierarchical structure and a device thereof are provided to determine a symbol encoding method considering a symbol feature such as symbol distribution in the video encoding device. CONSTITUTION: A video encoding device encodes a picture of video based on data units of a hierarchical structure(131). The video encoding device determines a symbol for the encoded picture(132). The video encoding device determines an encoding method of the symbol based on a determined symbol feature(133). The video encoding device encodes the symbol according to a determined encoding method(134).

Abstract translation: 目的：提供一种用于对分层结构中的符号进行编码/解码的视频编码/解码方法及其装置，以确定考虑诸如视频编码装置中的符号分布的符号特征的符号编码方法。 构成：视频编码装置根据层次结构的数据单位编码视频图像（131）。 视频编码装置确定编码图像的符号（132）。 视频编码装置基于确定的符号特征（133）确定符号的编码方法。 视频编码装置根据确定的编码方法编码符号（134）。

公开(公告)号：KR1020120008423A

公开(公告)日：2012-01-30

申请号：KR1020100097424

申请日：2010-10-06

Applicant: 삼성전자주식회사

Inventor： 민정혜 ,  이선일 ,  한우진

IPC: H04N19/105 ,  H04N19/51

CPC classification number: H04N19/593 ,  H04N19/105 ,  H04N19/11 ,  H04N19/14 ,  H04N19/167 ,  H04N19/176 ,  H04N19/182 ,  H04N19/196 ,  H04N19/513 ,  H04N19/80 ,  H04N19/91

Abstract: 영상의인트라예측부호화, 복호화방법및 장치가개시된다. 본발명에따른영상의인트라예측부호화방법은현재픽처를소정크기의적어도하나의블록으로분할하고, 현재블록내부의각 픽셀을중심으로소정의기울기를갖는연장선을이용하여이전에복호화된주변블록의픽셀들중 현재블록내부의각 픽셀의예측에이용될주변블록의픽셀을결정하고, 결정된주변블록의픽셀을이용하여현재블록내부의각 픽셀을예측하는것을특징으로한다.

Abstract translation: 目的：提供图像帧内预测编码和解码方法及其装置，以减少在帧内预测期间执行的操作量。 构成：帧内预测编码装置将当前图像划分为至少一个固定大小的块（1910）。 该装置确定要用于预测当前块中的每个像素的相邻块的像素，其中像素在先前解码的相邻块的像素之中（1920）。 该装置使用所确定的相邻块的像素来预测当前块中的每个像素（1930）。

公开(公告)号：KR1020110112187A

公开(公告)日：2011-10-12

申请号：KR1020110005003

申请日：2011-01-18

Applicant: 삼성전자주식회사

Inventor： 알쉬나엘레나 ,  알쉰알렉산더 ,  천민수 ,  한우진

IPC: H04N19/53 ,  H04N19/533 ,  H04N19/46 ,  H04N19/96 ,  H04N19/513 ,  H04N19/105 ,  H04N19/176 ,  H04N19/44

CPC classification number: H04N19/523 ,  H04N19/30 ,  H04N19/46 ,  H04N19/52 ,  H04N19/53 ,  H04N19/533 ,  H04N19/96 ,  H04N19/105 ,  H04N19/176 ,  H04N19/44 ,  H04N19/513

Abstract: 영상의 부호화, 복호화 방법 및 장치가 개시된다. 본 발명에 따른 영상 부호화 방법은 소정의 제 1 서브 픽셀 단위로 현재 블록에 대한 움직임 예측을 수행하여 현재 블록에 대한 제 1 서브 픽셀 단위의 움직임 벡터를 획득하고, 참조 픽처를 제 1 서브 픽셀 단위보다 작은 제 2 서브 픽셀 단위로 보간하며, 제 2 서브 픽셀 단위로 보간된 참조 픽처를 이용하여, 제 1 서브 픽셀 단위의 서브 픽셀의 주변에 위치한 제 2 서브 픽셀 단위의 서브 픽셀들 중 하나의 서브 픽셀을 선택하고, 움직임 벡터가 가리키는 참조 픽처의 제 1 서브 픽셀 단위의 서브 픽셀을 기준으로 획득된 제 1 대응 영역과, 제 2 서브 픽셀 단위의 서브 픽셀을 기준으로 획득된 현재 블록의 제 2 대응 영역 중 현재 블록과의 오차가 작은 대응 영역을 선택하고, 선택된 대응 영역의 정보를 부호화하는 것을 특징으로 한다.

公开(公告)号：KR1020110025444A

公开(公告)日：2011-03-10

申请号：KR1020090083512

申请日：2009-09-04

Applicant: 삼성전자주식회사

Inventor： 한우진 ,  박정훈 ,  정해경

IPC: H04N21/2389 ,  H04N21/4385 ,  H04N19/70 ,  H04N19/176 ,  H04N19/61 ,  H04N19/157 ,  H04N19/436 ,  H04N19/91 ,  H04N7/24

CPC classification number: H04N21/2389 ,  H04N19/157 ,  H04N19/176 ,  H04N19/436 ,  H04N19/61 ,  H04N19/70 ,  H04N19/91 ,  H04N21/4385 ,  H04N7/24

Abstract: PURPOSE: A method and apparatus for generating a bit stream based on a syntax element are provided to facilitate the storage and transmission of media data by compressing media through various encoding methods. CONSTITUTION: An encoding unit(110) generates image data including the data for syntax elements, and a grouping unit(120) groups the generated image data according to the kinds of syntax elements. A bit-stream generation unit(130) arranges the bit-stream of the image data in order. After the arrangement in order, the bit-stream generation unit adds a header to each group including information on the position of the header of the group.

Abstract translation: 目的：提供一种基于语法元素生成比特流的方法和装置，以通过各种编码方式压缩媒体来促进媒体数据的存储和传输。 构成：编码单元（110）生成包括语法元素的数据的图像数据，分组单元（120）根据语法元素的种类对所生成的图像数据进行分组。 比特流生成单元（130）按顺序排列图像数据的比特流。 在顺序排列之后，比特流生成单元将包括关于组的头部的位置的信息的标题添加到每个组。

公开(公告)号：KR1020090094595A

公开(公告)日：2009-09-08

申请号：KR1020080019642

申请日：2008-03-03

Applicant: 삼성전자주식회사

Inventor： 김일구 ,  한우진 ,  손유미

IPC: H04N19/573

CPC classification number: H04N19/517 ,  H04N19/105 ,  H04N19/147 ,  H04N19/154 ,  H04N19/176 ,  H04N19/46 ,  H04N19/523 ,  H04N19/56 ,  H04N19/61

Abstract: A method and an apparatus for encoding images using motion prediction by multiple reference, and a method and an apparatus for decoding images using motion prediction by multiple reference are provided to encode image data including a current frame based on a motion prediction performing result, thereby improving prediction performance between screens. A candidate estimator set determiner determines one or more candidate estimator sets including one or more candidate estimators. A reference estimator determiner(270) determines a reference estimator based on an error rate of each prediction estimator. A motion prediction performer(280) performs motion prediction about the current frame by using a reference block and the current block corresponding to the reference estimator. An encoding unit encodes image data including the current frame based on the motion prediction performing result.

Abstract translation: 提供了使用多参考的运动预测来对图像进行编码的方法和装置，并且提供了使用多参考的运动预测对图像进行解码的方法和装置，以基于运动预测执行结果对包括当前帧的图像数据进行编码，从而改善 屏幕之间的预测性能。 候选估计器组确定器确定包括一个或多个候选估计器的一个或多个候选估计器集合。 参考估计器确定器（270）基于每个预测估计器的错误率来确定参考估计器。 运动预测执行器（280）通过使用参考块和与参考估计器相对应的当前块来执行关于当前帧的运动预测。 编码单元基于运动预测执行结果对包括当前帧的图像数据进行编码。